Skip to main content

Full text of "Journal of the Society of Motion Picture Engineers"

See other formats


From the collection of the 

ft 

x z n m k 

o Prepnger h 
JJibrary 



v 



t 



San Francisco, California 
2007 



.UME XXIV NUMBER ONE 

JOURNAL 

of the 

SOCIETY OF MOTION 
PICTURE ENGINEERS 






JANUARY, 1935 



ISHED MONTHLY BY THE SOCIETY OF MOTION PICTURE ENGINEERS 



The Society of Motion Picture Engineers 
Its Aims and Accomplishments 

The Society was founded in 1916, its purpose as expressed in its 
constitution being the "advancement in the theory and practice of mo- 
tion picture engineering and the allied arts and sciences, the standardi- 
zation of the mechanisms and practices employed therein, and the 
maintenance of a high professional standing among its members." 

The membership of the Society is composed of the technical ex- 
perts in the various research laboratories and other engineering 
branches of the industry, executives in the manufacturing, produc- 
ing, and exhibiting branches, studio and laboratory technicians, 
cinematographers, projectionists, and others interested in or con- 
nected with the motion picture field. 

The Society holds two conventions a year, spring and fall, at various 
places and generally lasting four days. At these meetings papers 
dealing with all phases of the industry theoretical, technical, and 
practical are presented and discussed and equipment and methods 
are often demonstrated. A wide range of subjects is covered, many 
of the authors being the highest authorities in their particular lines 
of endeavor. Reports of the technical committees are presented and 
published semi-annually. On occasion, special developments, such 
as the S. M. P. E. Standard Visual and Sound Test Reels, designed for 
the general improvement of the motion picture art, are placed at 
the disposal of the membership and the industry. 

Papers presented at conventions, together with contributed arti- 
cles, translations and reprints, abstracts and abridgments, and other 
material of interest to the motion picture engineer are published 
monthly in the JOURNAL of the Society. The publications of the 
ty constitute the most complete existing technical library of 
the motion picture industry. 

Spring Convention 

Hotel Roosevelt, Hollywood, Calif,, May 20-24, 1935 

The Spring onvention will be held at Hollywood, May 20- 

Headquarters will be at the Hotel Roosevelt. An unusu- 

linical papers and presentations is being 

planned by Mr. J. I. Crabtree, Editorial Vice-President, and Mr. J. O. 
I J apers Committee. All members are urged to bend 
the Convention. 

. under the leadership of Mr. W. C. 

Kun/mann, >orating with the Board of Managers of the 

;oward making the Convention a most pleasant 



JOURNAL 



OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 

Volume XXIV JANUARY, 1935 Number 1 



CONTENTS 

Page 
Current Developments in Production Methods in Hollywood .... 

H. G. TASKER 3 

The Motion Picture Industry in the Soviet Union 

V. I. VERLINSKY 12 

Report of the Committee on Standards and Nomenclature 16 

Report of the Committee on Non-Theatrical Equipment 23 

Report of the Color Committee 29 

Report of the Historical and Museum Committee 31 

Report of the Projection Practice Committee 35 

The Non-Rotating High-Intensity D-C. Arc for Projection 

D. B. JOY AND E. R. GEIB 47 

The Development of 16-Mm. Sound Motion Pictures 

E. W. KELLOGG 63 

Society Announcements 91 



JOURNAL 

OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 



SYLVAN HARRIvS, EDITOR 

Board of Editors 
J. I. CRABTREE, Chairman 

O. M. GLUNT A. C. HARDY L. A. JONES 

J. O. BAKER 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts. f Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1935, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 



Officers of the Society 

President: HOMER G. TASKER, 4139 38th St., Long Island City, N. Y. 
Past-President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
Executive Vice-President: EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, 

Calif. 

Engineering Vice-President: LOYD A. JONES, Kodak Park, Rochester, N. Y. 
Editorial Vice-President: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Financial Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Convention Vice-President: WILLIAM C. KUNZMANN, Box 6087, Cleveland, Ohio. 
Secretary: JOHN H. KURLANDER, 2 Clearfield Ave., Boomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York, N. Y. 

Governors 

MAX C. BATSEL, Front & Market Sts., Camden, N. J. 
LAWRENCE W. DAVEE, 250 W. 57th St., New York, N. Y. 
ARTHUR S. DICKINSON, 28 W. 44th St., New York, N. Y. 
HERBERT GRIFFIN, 90 Gold St., New York, N. Y. 
WILBUR B. RAYTON, 635 St. Paul St., Rochester, N. Y. 
SIDNEY K. WOLF, 250 W. 57th St., New York, N. Y. 



CURRENT DEVELOPMENTS IN PRODUCTION METHODS 
IN HOLLYWOOD* 

H. G. TASKER** 



Summary. An informal and non-technical commentary on a number of com- 
paratively new aspects of Hollywood production methods. No attempt is made to 
include all the recent developments in this field, but examples are cited in set con- 
struction, lighting, photography, recording, re-recording, and processing. 

Undoubtedly a better title for this paper would be "Seeing Holly- 
wood with a Slide Rule," because most of the material was gathered 
during recent informal chats with engineering acquaintances in a 
number of Hollywood studios. I can offer no guarantee that the 
subject matter represents recent developments, or that the in- 
formation was given to me in each case by the engineer to whom I 
ascribe it, and still less assurance that the development, if such it be, 
was the work of that engineer. In many cases, however, these are the 
actual facts. I should add also that some of the production methods 
to be described are now in use in eastern studios as well as in Holly- 
wood. 

SCRIPT TIMING 

Perhaps we should begin this story at the beginning, when the 
young scenario, a bit awkward and overgrown, first reaches the 
studio and preparations are begun for its production. Even at 
this early stage in the life of a Hollywood production we find a neat 
bit of engineering being introduced at RKO under the direction of 
M. J. Abbott and later adopted by other studios. A group of now 
experienced script analysts study the proposed story to determine the 
playing time of each scene of the production and, from the total, 
determine how long the picture would run. It is at once shortened 
to proper release length before filming, thus avoiding the common and 
expensive procedure of throwing away several reels of completed 
negative in order to boil the picture down to acceptable length. 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 
** United Research Corp., Long Island City, N. Y. 



4 H. G. TASKER [J. S. M. P. E. 

They next determine the length of time each set will be occupied, 
to enable the studio manager to schedule the set construction properly 
on the various sound stages. It is quite often possible by a deft 
change of scene or story, to avoid the expense of building a set which 
might be used for only a half minute or so. 

The playing time of each actor is then investigated, with the re- 
sult that players may be engaged for just the right length of time to 
complete their parts of the picture. These and other data combine 
to make possible savings amounting to many thousands of dollars 
per feature. 

LIGHTING 

In company with Frank Murphy, genial impressario of lighting, 
mechanics, and almost everything else at Warner Bros. Burbank 
Studio, I hiked for several miles through shops and stages, and ob- 
served, among other items, how rhubidium mirrors, made by front- 
surface plating on copper backing, are eliminating the previous high 
toll for mirror breakage on incandescent studio lights. He showed me 
also a mile or more of hanging platforms >or catwalk structures, on 
which these lights are mounted. These platforms are obviously 
direct descendants from the ancient painters' scaffolds, and, like their 
forebears, are readily adjustable as to height and position, and hence 
are more convenient than the old semi-permanent hammer-and-nail 
construction. 

THE MECHANIZING TREND 

As our journey through the studio continued I could not help be- 
ing impressed by the trend toward mechanizing almost everything 
involved in making a motion picture except (I hope) the actors them- 
selves! Camera cranes and camera elevators, both large and small, 
swing camera men and cameras up and down, left and right, with al- 
most limitless freedom. Motorized sets perform in almost human 
fashion, their movements synchronized by those interlocking motors 
that were so long confined to synchronizing sound and picture ap- 
paratus. Of the latter, interesting examples are found in the many 
overgrown clock faces disappearing through the slotted floor in the 
picture Dames, in the symmetrically dividing steps of a musical 
picture now in production, and in the 16 belt-type tread-mills on 
which as many horses will race to victory or defeat in another picture 
soon to be produced. These tread-mills, one of which I saw com- 



Jan., 1935] PRODUCTION METHODS IN HOLLYWOOD 5 

pleted, are some 16 feet long and 4 feet wide, supported at either 
end by wheels 2 feet in diameter, which will be driven by the inter- 
locked motors. The surface of the tread-mill is made up of 3 by 3 
inch slats and is supported along its length by a track and rollers. 

The coming of sound sadly impaired the usefulness of the old 
wind machines, but until recently I have observed no substantial 
improvement in their construction. New machines are now in 
evidence with streamlined blades and bullet noses driven by quiet 
motors up to 25 hp. in size. They can hardly be called silent fans, 
but are a vast improvement. 

Beneath the floor of one large stage, I saw the swimming pool, 
some 50 by 75 feet in size, which was used in Footlight Parade. The 
sides and bottom of the pool are provided with windows through 
which lights or cameras are directed, and its water is cleaned by 
three filters, each some 6 feet in diameter and as many high. The 
pool has since been used in other pictures. 

PHOTOGRAPHY 

Al Tondreau, in charge of the camera shop at Warner Bros. Holly- 
wood Studio and originator of many gadgets for the convenient 
manipulation of cameras, showed me a new arrangement for varying 
the angle of the finder in accordance with changing focus of the photo- 
graphic lens so that the field-of-view of the finder always coincides 
with that of the picture lens in the plane of sharpest focus. Another 
device was a complete animation camera with many new gimicks 
and gadgets which was used, not for making Mickey Mouses, but 
for animation sequences in feature productions. 

Fred Jackman, in charge of the special effects department at 
Warner Bros., introduced me to process photography of the back- 
ground projection type in actual production. This method seems 
to have replaced the former color-separation method in Hollywood 
production, and consists in projecting the background scene upon 
one side of a translucent screen of any required size while the camera 
and local action are arranged on the other side. The shutters of 
projector and camera are synchronized by interlocking motors to 
avoid flicker. To obtain the needed brilliance of the projected pic- 
ture, high-density prints are avoided, and the arc lamp current runs 
as high as 225 amperes, the gate being cooled by a glycerine cell. 

This type of process photography is employed for other purposes 
than to save the railroad fares otherwise necessary to place actors 



6 H. G. TASKER [J. S. M. p. E. 

against more or less distant backgrounds. I was much interested 
to witness a take from Six Day Bicycle Race, in which a conversation 
took place between two riders supposedly tearing around the track 
at some 20 miles per hour. The difficulty of recording such a con- 
versation on an actual race-track, with both camera and microphones 
sweeping around the track ahead of the riders, may be easily ap- 
preciated. In the process shot the background consisted of a view 
of the race-track shot wild from the back of a motorcycle leading a 
group of racers by some 15 or 20 yards. In front of this projected 
image were the bicycles of the two conversing riders. These were 
supported on swiveled mounts which were swayed back and forth by 
stage hands, cleverly representing the movements required by the 
swirling race-track in the background shot. As the riders pedaled 
vigorously the front wheels of their bicycles were turned by small 
motors, and in front of the otherwise stationary riders a microphone 
was suspended which picked up the conversation between them. In 
this and many other instances I noted the continuing trend to push 
complexities of production into technical departments, to the great 
relief of shooting schedules. 

"ONE NIGHT OF LOVE" 

Having heard from many sources most interesting comments on 
the musical quality of Columbia's One Night of Love, starring Grace 
Moore, I searched out Mr. John Lividary, in charge of sound at that 
studio, and inquired of him as to how it was done. He was most 
cordial and informative, and at last convinced me that it was not 
done with mirrors. 

In recording the musical scenes two dynamic microphones were 
devoted to orchestral pick-up, one about 30 feet from the orchestra 
and a second 90 feet away. The latter was employed to introduce 
desired amounts of the reverberant quality that may have occasioned 
much of the popular enthusiasm for the final product. The stage 
being still too dead, even after introduction of much hard flat surface, 
a variable low-pass filter was associated with the second mike. Miss 
Moore desired unusually close association with the orchestra, and so 
the directional properties of a ribbon microphone, placed in front 
of Miss Moore and at right angles to most of the orchestra, were 
employed to provide satisfactory differentiation between voice and 
accompaniment . 

An added chorus called for another ribbon microphone, the bi- 



Jan., 1935] PRODUCTION METHODS IN HOLLYWOOD 7 

lateral properties of which permitted grouping the male voices on 
one side and the female voices on the other side in such numbers as 
would otherwise have required two mikes for such a chorus. 

The musical numbers were recorded simultaneously on film and 
on hill-and-dale disk, of which only the latter were used in the re- 
lease. The very wide volume-range of the hill-and-dale method en- 
abled successful "canning" of Miss Moore's none too predictable 
performance, which could thus be re-recorded with care and finesse 
so as just to reach, but never exceed, the upper volume limit of the 
release film. It was with some reluctance and only moderate success 
that the dialog level was lowered and the noise reduction increased 
to provide still further volume contrast between dialog and Miss 
Moore's musical numbers. 

All in all I am inclined to feel that the very considerable amount 
of favorable and even astonished comments on the sound quality of 
the finished picture is a tribute to the skill and the enterprise of an 
engineering organization that successfully adopted such unusual 
methods on what was practically their first musical production. One 
of the most important of these "unusual" methods was quite old in 
the art; namely, that of first recording on disk and later transferring 
to the release film. 

It would seem that portions, if not all, of the foregoing methods 
were destined to be adopted by other studios, and I learned from 
Loren L. Ryder, in charge of the finishing department at Paramount, 
that hill-and-dale disk recording had been used with excellent re- 
sults in connection with reels 1 and 7 of their forthcoming picture 
Enter Madame. In their use of hill-and-dale, Paramount has taken 
advantage of the wide volume-range to equalize the high frequencies 
upwardly to offset film-transfer losses by an amount not now possible 
with recordings originally made on film because of increased surface 
noise. 

DISK AND FILM PROCESSING 

The successful use of hill-and-dale recording in these examples 
caused me to inquire of Kenneth Morgan and A. P. Hill, of Electrical 
Research Products, Inc., as to what had become of the previously 
too familiar "pops" or "crackles" of the hill-and-dale recordings. 
For answer they took me to the Hollywood laboratory of the company. 
There they first let me listen to a number of completely "popless" 
and very excellent hill-and-dale records, and then showed me the 



8 H. G. TASKER [J. S. M. P. E. 

dust-free and air-conditioned portion of the laboratory where the 
processing of the records is carried on. Complete freedom from 
dust throughout the operation from sputtering to pressing is said 
to be the principal key to successful elimination of the "pops." 

In the field of film processing I found steps being taken to mini- 
mize the directional effects caused by the restraining effects of re- 
action products of development. To the usual jets are being added 
mechanical agitators, which are expected to reduce this annoying 
source of wave-form distortion to negligible proportions. 

CELLULOSE DISK PLAYBACK 

It is well known that large musical numbers are customarily pre- 
scored either on disk or film, and the sound record thus produced 
is employed to rehearse and finally photograph the movements of 
the players. Disk records have been most commonly used, but or- 
dinarily involve a time lapse of six hours to two days for the return of 
a finished record which can be played often enough for these re- 
hearsals. From G. M. Best, of Warner Bros, sound department, I 
learned of the rapidly growing use of recordings directly cut upon 
a cellulose disk. Although the material used is somewhat similar 
to that employed for home recording a few years ago, the method 
differs essentially in that the record surface is actually cut away as 
in the former wax-recording, rather than being embossed into a pre- 
grooved blank as in the home-recording method. 

Stimulated by the need for more rapid preparation of playback 
records the development of this type of recording has progressed 
to the point where the sound quality is very satisfactory and the 
surface noise substantially lower than in film recordings or the con- 
ventional disk. Each record may be played as many as one hundred 
times, and is used on the set in full sight of the players so that the 
record may be started, stopped, or re-started at the nod of the di- 
rector or artist. 

These new records are almost instantly available for use. In a 
typical instance, eight minutes elapsed between the beginning of a 
playback recording and the time the completed playback record was 
delivered to the stage and placed in operation. 

Cellulose materials available are of great variety : some solid cellu- 
lose, some coated on steel or aluminum cores, some clear, some 
opaque, and, I might add, some good and some bad! In the best in- 
stances they have such physical properties that sapphire cutting 



Jan., 1935] PRODUCTION METHODS IN HOLLYWOOD 9 

tools may be employed with a recording life of 2 1 / z hours and very 
excellent results. 

All the Busby Berkeley dance numbers, several of which I saw 
in preparation, are prescored, rehearsed, and photographed with 
these records directly cut on cellulose disk. A single number is often 
played as many as two hundred or three hundred times, in which case 
two or more disks are provided either by direct or re-recording. 

SOUND COORDINATION 

During a conversation with Colonel Nugent H. Slaughter, of 
Columbia, and Wallace B. Wolfe, of General Service Studio, Inc., 
the latter raised a technical issue that is an important matter for 
S.M. P.E. consideration. Mr. Wolfe expressed the opinion that 
something should be done promptly to bring about coordination of 
the product of the various studios as to sound quality, particularly 
with respect to frequency characteristic and output level. In these 
respects there is still a very wide variation, resulting in much need- 
lessly poor performance in the theater. I was pleased to report to 
him that the S.M. P. E., through its Sound Committee, L. W. Davee, 
Chairman, was actively studying this problem. A most important 
initial step, the establishment of a quality yard-stick, has already been 
undertaken and I have reason to believe that a workable basis for 
studio coordination will result. The Hollywood section of the 
Sound Committee will doubtless find studio engineers vitally in- 
terested. 

FINISHING DEPARTMENT COORDINATION 

As in most other industries, probably the most interesting depart- 
ment of motion picture production is the assembly plant, where the 
pieces of the "jig-saw puzzle" are finally fitted together. These 
operations in most studios are the responsibility of the Sound De- 
partment. The manner in which these departments gather to- 
gether the needed information and material to complete the finished 
product varies widely from studio to studio, as described by Colonel 
Slaughter, of Columbia, Loren Ryder, of Paramount, and Chester 
North, of Warner Brothers, in charge of their respective finishing de- 
partments. In some studios the finishing department reads the 
script and consults with the director early in the course of the pro- 
duction, to determine upon methods of recording or re-recording to 
obtain the desired results. In others the material is practically 



10 H. G. TASKER [J. S. M P. E. 

completed before the finishing department appears in the picture. 
In some studios needed accessory sounds or effects are gathered while 
the production is under way; in others the directors object to this 
practice and sound-effects are separately created by crews assigned 
to that duty; and still other studios depend largely upon vast 
libraries of sound-effects with few additions of new material. In 
some studios all the sound is re-recorded for the proper blending-in 
of sounds not present during the original take, or for other modi- 
fication ; while in others as little as half the master negative is derived 
by re-recording. 

The tools of the finishing department have advanced with the art, 
and notable among them is a device shown to me by William A. 
Mueller, in charge of sound equipment at Warner Brothers, which 
automatically regulates music and dialog in certain scenes. This 
device, popularly known as the "up-and-downer," causes the level 
of re-recorded music to be automatically lowered word by word in 
response to dialog so that the latter is clearly heard above the music, 
while permitting full volume of music to occur at all other times. A 
very useful example occurs when conversation takes place between a 
couple on a dance floor. 

"MUSICAL MOODS" 

As we leave the studio, our Hollywood production safely in the 
shipping room, it will be interesting to note a bit of "program en- 
gineering" which was brought to my attention while visiting the 
Hollywood Laboratory of Electrical Research Products, Inc., as the 
guest of Kenneth Morgan and A. P. Hill, of that organization. Not- 
ing with some concern the intensely swift tempo of the entire motion 
picture program as presented in the average theater, Electrical Re- 
search Products, Inc., have felt that it might be better balanced and 
more thoroughly appreciated if some program element could be intro- 
duced to slow the pace a bit and relax the audience tension. Ac- 
cordingly they have created a group of numbers known as "Musical 
Moods," of which it was my pleasure to observe one, entitled In a 
Monastery Garden. This picture is a delightfully restful yet very 
interesting combination of good music and beautiful scenes in 
Technicolor, for which I feel the originators are to be congratu- 
lated. 



Jan., 1935] PRODUCTION METHODS IN HOLLYWOOD 11 

DISCUSSION 

MR. RICHARDSON: We often have this situation in the theater: Projection 
is progressing very well, with the sound at its proper level. Suddenly the sound 
decreases almost to nothing, or increases to far above its normal level. The dialog 
may be lost, or the ears of the patrons strained. In either case, it is annoying. 

Such a situation is bad enough in two-man projection rooms; but in one-man 
rooms it is often beyond control, because other duties that can not be neglected 
keep the projectionist away from his machine. What are the causes of such 
situations, and how can they be remedied? 

MR. TASKER: Certain engineers are inclined to feel that responsibility for 
most of the trouble that you describe lies at the door of the laboratories, due to 
inconsistency in processing the film. I am told that upon measuring certain 
release films that were said to have gone through the printer at uniform light 
intensity and should also have gone through the developing bath with uniformity, 
density changes were discovered in the unmodulated portion almost too great to 
measure. I am told, however, that not all laboratories are guilty in that respect 
and that there are a number of plants that are doing a really fine job. 

MR. EVANS: A year or so ago I proposed a scheme that the Sound Committee 
thought the Society was hardly in a position to inaugurate. I was not quite con- 
vinced that that was a fact. 

' Two factors determine the reproduction level: One is the percentage of modu- 
lation in recording, and the other is the density (or transmission) of the print. 
It is impossible to maintain the percentage of modulation exactly the same from 
day to day and under all recording conditions. Variations hi recording, however, 
can be, and are, corrected by re-recording. 

It is not possible when printing in large quantities to make the density of 
all prints exactly the same. It is possible, however, to sort the reels so that the 
variation in density in any set of reels will cause less variation in level than one 
fader step on the reproducer. While that has been done, and is being done, it 
does not seem to me to be a very satisfactory solution of the problem. 

My suggestion is that each laboratory determine the correct reproducing 
level for prints of normal density (transmission) from each negative. With the 
aid of a density or transmission table the correct reproducing level could then be 
determined for prints of all other densities from the same negative. 

A place should be reserved on the standard Academy leader where the correct 
reproducing level, in decibels above or below normal, should be stamped. Each 
theater has (or should have) a normal reproducing level for each machine, and 
by looking at the level correction stamped on the leader of each reel, the operator 
could determine how he should set the volume control for that particular reel. 

That would not be a difficult or expensive procedure for the laboratories to 
follow, and it would do much toward standardizing reproduction levels in theaters 
and make it possible for careful operators to eliminate practically all the level 
changes that now occur in changing over from one reel to the next. 



THE MOTION PICTURE INDUSTRY IN THE SOVIET UNION* 

V. I. VERLINSKY** 



Summary. The growth of the motion picture industry in the Soviet Union during 
the past ten years is briefly traced; then the troublesome transition period following 
the advent of sound; and finally the status upon which the present motion picture 
activities are organized, leading to the production and distribution of highly artistic 
features, newsreels, scientific films, etc. 

The motion picture industry of the Soviet Union is essentially a 
growth of the past ten years, but, if I may say so, a substantial and 
healthy growth. During the czarist regime the industry was insig- 
nificant, most of the films being imported from foreign countries. 
When the Soviet Government came into power in 1917, it found itself 
in possession of two small studios which represented the entire pro- 
ductive equipment of the country. 

Conditions during the next few years were not conducive to the ex- 
pansion of the industry. The few films made during this period were 
documentary records of important events during those stormy days, 
perhaps of no great artistic pretensions, but of increasing historical 
importance. It was not until the close of the period of civil wars and 
invasions that the country was in a position to build up the motion 
picture industry. During the past decade the growth has been rapid. 
Today there are over ten studios, located in Moscow, Leningrad, 
Odessa, Kiev, Yalta, and other centers, and the country has become 
one of the most important world-producers. There are upwards 
of 30,000 theaters, and the attendance at performances was over 
650,000,000 last year. In addition to motion picture schools for the 
development of its artistic and technical staffs, the industry has its 
own academy and a special research institute. 

The days of the silent pictures brought to the fore some great 
masters and some great productions, which commanded artistic 
admiration and respect far beyond the Soviet borders. Eisenstein's 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 
** Amkino Corp., New York, N. Y. 
12 



MOTION PICTURES IN SOVIET UNION 13 

Potemkin and Pudovkin's Mother were among the pioneers in these 
great works, and others outstanding in this period included Dov- 
zhenko's Soil and Arsenal and Pudovkin's Storm over Asia and End 
of St. Petersburg. 

The transition from silent to sound pictures has been a peculiarly 
difficult one in the Soviet industry for many reasons, not the least 
of which is the fact that the Soviet Union is composed of 182 dif- 
ferent nationalistic stocks speaking 150 different languages and 
dialects. Obviously the creation of talking pictures for such a 
polyglot population presents special problems. 

Within a short time after the coming of the talking pictures, the 
Soviet industry produced a masterpiece in the new form, The Road 
to Life, which won international acclaim. It had no immediate suc- 
cessors of similar caliber. The adjustment to the new medium 
brought a period of groping and of struggle, and what finally emerged 
was essentially a collective product, based upon a realism grounded 
on life within the socialistic structure of the country, and the style 
of it is termed "socialistic realism." An excellent example of this 
emergent form was the film Shame, made by Ermler and Youtke- 
vitch, which dealt with the life and problems of the Soviet citizens 
engaged in collective work. The diversity and resourcefulness in 
subject, treatment, and technic are shown also in the following sound 
productions : the historical film Thunderstorm, taken from Ostrovsky's 
play and directed by Petrov; Petersburg Nights, based on Dos- 
toyevski's novel and directed by Roshal and Stroyeva; Three Songs 
about Lenin by Vertov ; the film of the Arctic voyage of the Chelius- 
kin by Shafran ; the first musical jazz comedy by Alexandrov, and the 
first full-length feature with animated dolls, Gulliver's Travels. In 
addition, the Soviet industry is making its first full-length color 
film, Nightingale, by Ekk, director of the Road to Life. The Soviet 
studios are working beyond their capacity. For 1935, 150 full- 
length pictures will be produced, in addition to many short subjects 
and newsreels. The most recent efforts of the Soviet film industry 
tend toward a mastery of film technic resulting in a finished product 
of great art. Such films have already been made and have received 
universal recognition, as was shown at the International Motion 
Picture Exhibition in Venice, where the Soviet Union was awarded 
first prize as the producer of the world's most artistic films. 

The motion picture industry of the Soviet Union has been developed 
under the aegis of the Government. Each of the seven constituent 



14 V. I. VERLINSKY [J. s. M. p. E 

Republics of the Union has its own motion picture industry, operat- 
ing under the People's Commissariat for Education of the Republic 
in which it exists. The whole industry is combined in the Motion 
Picture Trust of the U. S. S. R. It is the problem of each division 
of the industry to satisfy the public in its territory. 

All the silent films produced have to be released with titles in some 
150 languages, to accommodate the entire polyglot population of 
the Soviet Union. The talking pictures are made in ten principal 
languages, and have superimposed titles for the various minor 
linguistic groups. In this respect the Soviet industry is faced by a 
complication that does not affect the industry in the United States. 

All the newsreels in the Soviet Union are under the control of the 
newsreel trust, Soyuzfilmnews, which takes care of the entire ter- 
ritory of the U. S. S. R. Every month Soyuzfilmnews issues three 
silent newsreels of general interest, three sound newsreels of general 
interest, two shorts on village life, a special short devoted to children, 
one on science and mechanics, one on art, and one on national de- 
fense. In addition Soyuzfilmnews participates in all scientific ex- 
peditions, and in this line has produced films of such expeditions as 
those of the Sibiryakov and the Cheliuskin and the exploration of the 
desert Kara-Kum. The trust has over 100 news cameramen scat- 
tered over the Soviet Union. The aim of Soyuzfilmnews is to install 
a system similar to that of the American newsreel companies to en- 
able them to have the newsreels in the theaters 24 hours after being 
filmed. 

All the scientific films are produced by a special scientific trust, 
which is assisted by the leading scientists, those of the Academy of 
Science of the U. S. S. R. including the famous physicist Pavlov. In 
1933 the trust issued 107 silent short subjects and only three sound; 
in the first nine months of 1934 there were 177 silent short subjects 
and 24 sound. 

It is only lately that the Soviet Union has begun to develop its 
own manufacture of raw-stock and equipment. One of the factories 
operating at the present time delivers tens of millions of film foot- 
age per year, but it is unable to meet the demand of the rapidly ex- 
panding industry. A new large factory is being built at Kazan, 
at a cost of 15,000,000 rubles, which will turn out over 300 million 
feet of raw-stock per year. Under the supervision of two Soviet 
sound inventors, Professor Chorine (who is a member of the S. M.- 
P. E.) and Mr. Tager, a large factory and laboratory is being operated 



Jan., 1935] MOTION PICTURES IN SOVIET UNION 15 

which supplies all the sound projectors for the theaters and conducts 
important research work for improving sound recording. The dis- 
tribution of the entire film output is in the hands of a special dis- 
tributing trust, which buys all the products from the producers and 
distributes them through its branches covering all the theaters of 
the Soviet Union. 

The number of motion picture theaters has shown a fifteen-fold 
increase during the past nine years. In 1924 there were 1953 the- 
aters; in 1932 there were 29,163. About 2000 are equipped for 
sound. The following table shows the growth of attendance at 
motion picture houses. 

1928 1932 

In the cities 233,270,000 447,722,000 

In the country 6,790,400 188,960,000 



Total 240,060,400 636,682,000 

The Soviet motion picture industry has a foreign department which 
takes care of selling Soviet pictures abroad and arranges for the pur- 
chase of foreign pictures and equipment. This department has its 
own representatives in Paris and in New York (Amkino) and has 
agreements with firms in several foreign countries. The foreign 
department has recently closed a deal in Paris with one of the major 
groups of European distributors for exclusive rights of distribution 
of all Soviet pictures in Europe, with a reciprocal provision for pro- 
duction of European films in the Soviet Union. This mutually 
beneficial contract may serve as a model for similar distributing ar- 
rangements in the United States. Amkino, it may be added, is 
interested in buying American films suitable for the Soviet market. 

The Soviet film industry has the highest admiration for the ad- 
vanced technic of American production. It offers a market for 
American machinery and chemicals used in the industry, and has an 
interest in securing the services of American experts to assist in its 
technical development. I trust that the friendly ties which exist 
between the American and the Soviet motion picture industries may 
continue and grow stronger with the passing years. 



REPORT OF THE COMMITTEE ON STANDARDS AND 
NOMENCLATURE* 

The activities of the Standards Committee since the Spring Con- 
vention were restricted principally to the completion of the Standards 
Booklet, which has already been published in the November, 1934, 
issue of the JOURNAL, and to the consideration of proposals from 
Europe relating to the standardization of 16-mm. sound-film. It 
was previously stated 1 that a report from the German Standardizing 
Committee would receive the attention of the S. M. P. E. Standards 
Committee and that an official reply would be prepared. 

The comments on the German report were forwarded to Mr. M. 
Flinker, Chairman, on June 7, 1934, to which the reply is as follows : 

(/) With reference to the difference between the German proposals and the 
American standard regarding the sound-track position relative to the picture, it 
would seem that the advantages as listed in favor of the German proposals would 
only result from certain specific mechanical designs, and that exactly the opposite 
conclusions hi regard to advantages can be reached if other specific mechanical 
designs are contemplated. 

The advantages of putting the sound-track on the inside that have been found 
from actual experience with machines commercialry manufactured by three 
companies in this country, are as follows : 

(1) Because of the desirability of making the apparatus small and compact, 
the photoelectric cell can be much more easily shielded electrically and 
from general illumination quite often coming from a-c. light sources that 
produce hum. 

(2) The lamp and optical system are mounted closer to the main supporting 
frame, assisting in the elimination of vibrations in these parts; vibration 
of the optical parts being, of course, just as detrimental as vibration of 
the parts carrying the film. 

(5) In equipment of this type, experience has shown that the sound optical 
system is much more easily adjusted and focused by aural methods than 
visually, so that the ability to observe the light-beam is of little practical 
consequence. 

(4) It is quite practicable to design the parts so that cleaning of sound-track 
guides and optical parts is convenient with the sound-track on either side. 

(5) The facility with which accumulated dirt or film particles can be removed 
is entirely dependent upon the design. 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 
16 



STANDARDS AND NOMENCLATURE COMMITTEE 17 

(6) We do not understand the reason for the statement that with the perfora- 
tions on the inside, a greater economy in the film movement can be ob- 
tained, it being just as important, if not more so, to maintain freedom 
from eccentricity or inaccuracy of movement of the film at the point of 
sound reproduction as at the point where the perforations engage the 
sprocket teeth. If anything, it is preferable to scan the film at a point 
near the shaft bearings than to have the engaging sprocket teeth at this 
position. 

(7) The American Standards Committee has not concluded that it is impor- 
tant to have the perforations on the inside to facilitate threading and to 
prevent scratching of the sound-track, because threading is usually done 
on blank leader where scratching would be of little consequence, and also 
the threading is no more complicated than it has been for doubly per- 
forated film. Since there are no sprocket teeth adjacent to the sound- 
track, there would be little likelihood of scratching the sound-track when 
locating the perforations on the teeth. 

There have already been sold in the United States by three manufacturers ap- 
proximately 1000 projectors. There has been no indication that the sound-track 
arrangement with respect to the picture is unsatisfactory. Quite an extensive 
library of films has been established and many publicity and educational pictures 
have been made. 

(II) As regards the position of the emulsion side in the projector, the American 
Standards Committee is in complete agreement with the German Committee in 
that it is not possible in every case to establish for all types of film a uniform 
position of the emulsion side in the projector. The following considerations have 
governed in considering the necessity for a differential focusing of the sound optical 
system : 

(1) The largest field of use for 16-mm. sound equipment in the American 
market is believed to be in the use of the equipment as an entertainment 
medium in the home. The amateur or private user of this equipment 
has two sources of film : 

(a) Film which may be made on an amateur sound recording camera on 
reversible stock. 

(6) Films obtained from a rental library which would be almost univer- 
sally produced by optical reduction from 35-mm. pictures. 

As stated in the German report, pictures made by original exposure, 
picture and sound on 16-mm. reversal film, should be projected 
with the emulsion side toward the projection lens. 

(2) Reductions from 35-mm. negative stock to 16-mm. film and optical print- 
ing from original 16-mm. exposures: 

The opinion of the Standards Committee and laboratory technicians con- 
sulted is that there is no advantage in optical printing with the two emul- 
sion sides facing one another. There would seem to be no objection to 
printing through the support, and it is noted that all theater projection is 
carried out in this manner. Since rental libraries' film will be produced 
by optical reduction, there is no objection technically to making them 



18 STANDARDS AND NOMENCLATURE COMMITTEE [J. s. M. p. E. 

conform to the American standard. It is obvious that library films and 
films produced by the amateur or home movie-maker can be projected 
with the emulsion toward the screen in both cases and without having 
to refocus the sound optical system. 
(5) Original exposure on color-film according to the Keller- Dorian process: 

The American Committee agrees with the German Committee that this 
film must be projected with the emulsion side toward the light source. 
This statement also applies to contact prints made from original 16-mm. 
negatives. Colored pictures by the Keller- Dorian process have not been 
used extensively in the amateur field, and contact prints very little. 
Most users of the contact prints use them for some special application 
in which the machine will be operated almost exclusively for this type of 
print, and the sound optical system can be focused for such prints without 
inconvenience or the necessity for refocusing it often. 

A note is being added to the drawing of the American standard to the 
effect that color-film by the Keller-Dorian process, and contact prints 
made from original 16-mm. negatives, are to be projected with the emul- 
sion toward the light source. 

The reply to the German report expressed regret that it had been 
impossible to reply officially at an earlier date. 

On June 22, a meeting was called by the Educational Branch of 
the League of Nations, of which Dr. DeFeo, General Secretary of the 
International Educational Cinematographic Institute, was Chair- 
man, at Stresa, Italy, for the purpose of recommending to the League 
of Nations a standard 16-mm. sound film for use in the educational 
field. On very short notice, the Society of Motion Picture Engineers 
was invited to send a representative to the meeting. Dr. Goldsmith, 
President of the Society, appointed Mr. W. F. Garling, of the London 
office of an American manufacturer, and a member of the Society, as 
the S. M. P. E. delegate to the meeting. While this meeting was called 
for the purpose of recommending to the League of Nations a sound 
film for educational use, a resolution was passed recommending that a 
standard be adopted in accordance with the German proposal, and 
the Society of Motion Picture Engineers was requested to consider 
changing the American standard to conform to the German proposal. 
It was stated that unless action were taken by the Society of Motion 
Picture Engineers within six weeks, it was decided by the Committee 
meeting at Stresa that the German proposal would become standard 
in all countries holding membership in the League of Nations for 
16-mm. educational films. 

When a complete report of the meeting held at Stresa was received 
by the S. M. P. E., a meeting was called of the Standards Committee 



Jan., 1935] STANDARDS AND NOMENCLATURE COMMITTEE 19 

to review the minutes of the Stresa meeting, and to give consideration 
to the proposals made. As a result of the discussion and opinions ex- 
pressed at the meeting, the Standards Committee Chairman sent a 
letter report to Dr. Goldsmith. Doctor Goldsmith advised Dr. 
P. G. Agnew, Secretary of the American Standards Association, of 
the views of the S. M. P. E. Standards Committee by letter of 
July 30. 

Under date of August 10, Dr. Goldsmith officially replied to 
Dr. DeFeo's letter which advised the S. M. P. E., of the action taken at 
Stresa. His letter which follows was based on the action taken by 
the Standards Committee: 

In this communication you request that we ratify the agreement reached at 
Stresa. We regret that we are unable to do so, and desire to place before you the 
following facts: 

It is noted from the communication that the Committee was instructed to limit 
the discussion to two considerations: 

(1) They should study the possibility of the use of the sub-standard film for 
educational films before considering the use of this size for the amateur. 

(2} They should examine the technical evidence placed before them, and that 
they put forward a solution of the problem of standardization of sub- 
standard film for educational use. The Committee was requested to 
confine then- deliberations to this problem, leaving out of their considera- 
tions the question of other applications. 

A careful review of the minutes of the meeting enclosed with your letter leads 
to the conclusion that the discussion brought out the fact that there is no basis 
for a technical choice in regard to the positioning of the sound-track on one side 
of the picture with respect to the other. There was nothing that was brought out 
in the discussion, as shown in the minutes, that would seem to justify the resolu- 
tion in favor of placing the sound-track in accordance with the DIN proposal. 

The facts are that the German Standards Committee believed that they were 
adopting the S. M. P. E. standard when their proposed standard was issued. Since 
the drawing prepared by the German Standards Committee differed from the 
S. M. P. E. drawing because of a mistake in interpreting the S. M. P. E. drawing, and 
since the American standard was adopted several months before action was taken 
by the German Committee and equipment manufactured in America prior to 
manufacture in other countries, it seems reasonable that these facts should be 
taken into account. 

It is evident that it is unnecessary to establish a new and different standard 
for the educational work of the League of Nations. A second standard will lead 
to the usual complications and annoyances that are sure to result if the scope of 
activity is to be world-wide. 

The S. M. P. E. standard is the only one so far as we can determine that is used 
internationally, and in accordance with which equipment has been manufactured 



20 STANDARDS AND NOMENCLATURE COMMITTEE [J. S. M. P. E. 

in more than one country. It has been adopted in England by several concerns, 
and also by one or more in France, and was intended for adoption in Germany, 
and would have been adopted except for an unfortunate misinterpretation of 
drawings published in the S. M. P. E. JOURNAL. This standard has been found 
satisfactory by the concerns having had experience with it. No suggestions 
which would render it more suitable for the educational field have been received. 
It is suggested for your application that you give serious consideration to the 
possibility of adopting the S. M. P. E. standard for the I. C. E. 

It is believed that if the Committee had approached the problem from the 
standpoint of determining whether or not the S. M. P. E. standard was satis- 
factory, they would have been compelled to find that it was. Such a decision 
would, apparently, have resulted in an international standard without the neces- 
sity for further discussions, and would have fixed a standard for all applications of 
16-mm. sound film. As it is, it is believed that this general subject is still one to 
be acted upon by the official standardizing agencies of each country involved. 

The American industry has decided that it is not possible to accept a com- 
promise of the S. M. P. E. standard. We are extremely sorry that two 16-mm. 
standards will exist, because of the complications and additional expense which 
may result for those making use of 16-mm. sound film internationally. 

With specific reference to your letter of July 17, the friendly spirit of which 
we appreciate, I may say that we are fully in accord as to the desirability of a 
single and universally accepted 16-mm. dimensional standard for sound-on-film. 
Indeed, we believe that the progress of this art and the effectiveness of the ap- 
plication of these films to educational uses in Europe and America may, in large 
measure, depend upon such general agreement and standardization. It is for this 
reason that we particularly regret the unacceptable nature of the findings of the 
Stresa Conference from the viewpoint of the well-established and thoroughly 
experienced group of American manufacturers of sound-on-film equipment and 
from the viewpoint of the technical committees of the Society of Motion Picture 
Engineers. We shall be obliged if you will notify all concerned in the matter 
that the American motion picture industry and the Society of Motion Picture 
Engineers are not able to accept the proposals of the Stresa Conference, but will 
adhere to the present widely used standards of the Society of Motion Picture 
Engineers. 

Dr. DeFeo has replied to Dr. Goldsmith stating his regrets and that 
he had forwarded a reply to the interested parties, and pointed out the 
disadvantages of two standards. Dr. Goldsmith has replied to this 
letter stating that none of the considerations advanced by Dr. DeFeo 
justified the reconsideration of the subject. 

The Standards Committee appreciates the cooperation received 
from the American Standards Association, the various manufacturing 
concerns interested in 16-mm. sound equipment and films, and also 
from Dr. Goldsmith who has devoted considerable time to the neces- 
sary correspondence. 



Jan., 1935] STANDARDS AND NOMENCLATURE COMMITTEE 21 

M. C. BATSEL, Chairman 

W. H. CARSON H. GRIFFIN C. N. REIFSTECK 

E. K. CARVER A. C. HARDY H. RUBIN 

L. E. CLARK R. C. HUBBARD H. B. SANTEE 

J. A. DUBRAY L. A. JONES V. B. SEASE 

P. H. EVANS N. M. LAPORTE J. L. SPENCE 

R. M. EVANS C. W. LOOTENS E. I. SPONABLE 

R. E. FARNHAM D. MACKENZIE H. M. STOLLER 

C. L. FARRAND G. F. RACKETT S. K. WOLF 

W. B. RAYTON 

REFERENCE 

1 Report of the Committee on Standards and Nomenclature, /. Soc. Mot. Pict. 
Eng., XXXIII (July, 1934), No. 1, p. 3. 

DISCUSSION 

PRESIDENT GOLDSMITH: It is desirable to explain briefly two considerations 
that guided the Society in its correspondence with Dr. L. De Feo (who called the 
Stresa Conference which resulted in confusion and difficulty for us). The Society 
does not believe that it can do other than "play fair" with the industry and the 
workers in this field in the United States; in other words, since 16-mm. sound 
film was, if not entirely, at least in considerable measure, pioneered in the United 
States, and since the corresponding standards were thoughtfully and painstakingly 
worked out, and since the industry adopted those standards after mature con- 
sideration, and in good faith built equipment and provided manufacturing facili- 
ties for them, and since publication was arranged and approved on the subject 
of the dimensions, and since millions of feet of film were produced according to 
those standards, we believed that for no light reason would we be justified in 
throwing our conclusions overboard and converting into junk valuable pro- 
jectors and film. We should have to be persuaded by unanswerable technical 
arguments that there was a sheer and compelling necessity for so doing. 

We were, however, not so persuaded; and we have in all sincerity taken a 
firm stand that we shall not change unless and until some conclusive and persuasive 
reason for changing shall have been brought to our attention. 

In the second place, we objected to educational groups as such, standardizing 
dimensional standards for film just as we should have objected to industrial 
groups as such or entertainment groups as such. We felt that it was an absurdity 
that such specialized standardization should proceed. In other words, if there 
is to be a 16-mm. sound-film standard for education, it should in our opinion be 
identical for industrial films or entertainment films. As Mr. Jones has well put 
it, the whole problem of 16-mni. sound-film standards should be considered at 
one time. 

In the third place, we found certain unusual elements in the entire procedure 
at Stresa. The International Educational Cinematographic Institute, which 
Dr. De Feo heads in Rome, is an interesting organization. We do not know its 
exact relation to the League of Nations. We saw no reason, however, why the 
Institute should set itself up as an international technical standardizing body, 



22 STANDARDS AND NOMENCLATURE COMMITTEE 

particularly on sound-film dimensional standards. This Society of Motion 
Picture Engineers has a Standards Committee which is broadly representative. 
It has prepared standards and these standards will now be handed over to a new 
Committee, the Sectional Committee on Motion Pictures, which has been author- 
ized by the American Standards Association, under the sponsorship of the Society 
of Motion Picture Engineers. 

I know you will be pleased to hear in that connection that the American Stand- 
ards Association, which validates and issues American standards, has designated 
the Society of Motion Picture Engineers as the sponsor of this new Sectional 
Committee on Motion Pictures, which will handle broadly all branches of motion 
picture standardization and will have on it full representation of all important 
individuals, groups, and interests in the motion picture industry who may be 
interested. 

We believe that standards should pass through our Standards Committee to 
this new Sectional Committee, and after validation they will then become Ameri- 
can standards. We believe that those standards should go through the American 
Standards Association to the International Standards Association if they are to 
be validated internationally. We are opposed to special procedures whereby 
particular bodies claiming to cover sections of the motion picture field shall, 
without our request or authority, and outside the regular procedure described 
above, allocate to themselves the task of dimensional standardization. It is 
our present intention to refrain from participation in such activities, since by 
implication we might otherwise lend a sanction to them. 

I am laying these considerations before you in order that the membership of 
the Society may know what action the Board and the President have taken on 
your behalf during the past year in connection with this confusing and unfortunate 
matter. Howevsr, so far as our American standards are concerned, we have 
endeavored to keep our procedure perfectly clear and fair to all concerned. 



REPORT OF THE NON-THEATRICAL EQUIPMENT 
COMMITTEE* 

This report is largely supplementary to the general survey pre- 
sented at the last convention. 1 The intervening period has shown a 
great increase in the use of sound film for advertising and educational 
purposes. While a larger portion of the films are on 16-mm., one 
manufacturer reports a large increase in the use of 35-mm. portable 
projectors. 

As may be gathered from a following section of this report, the use 
of 16-mm. films, both silent and sound, for advertising purposes re- 
ceived a definite impetus at the World's Fair at Chicago. At least one 
large motor car manufacturer has allocated the major part of this 
year's advertising appropriation to talking pictures. The pictures 
are shown throughout the country on 16-mm. sound outfits, and the 
success of the undertaking has aroused extreme interest not only 
among other motor car manufacturers, but among manufacturers in 
other lines. It seems inevitable that as a result of this successful 
application, the use of 16-mm. talking films for advertising purposes 
will spread rapidly abroad. 

Impetus has been given to the use of 8-mm. film by the introduc- 
tion of a powerful gear-driven projector utilizing a 300-watt lamp, 
enabling brilliant pictures as large as 5 to 7 feet in size to be projected 
from the small film. Since the last report, 750- watt lamps have 
become well established for 16-mm. projectors, especially with 
Kodacolor, and this month a 1000-watt projector was announced as 
being available soon. It has been announced that at a semi-private 
showing of this machine a picture 16 X 20 feet in size, of theatrical 
brilliance, was shown. 

From the foregoing it will be seen that 35-mm. films are being 
used more extensively for non-theatrical purposes; and on the other 
hand, the newer 16-mm. projectors seem more and more likely to 
invade the theatrical field or at least, the semi- theatrical field. 
This invasion has already taken place to some extent with "spon- 
sored" shows, church and similar organization fund-raising shows, 
and in the lecture field. 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 



24 NON-THEATRICAL EQUIPMENT COMMITTEE f J. S. M. P. E. 

A FEW FACTS REGARDING NON-THEATRICAL MOVIES AT THE CENTURY OF 
PROGRESS EXPOSITION IN CHICAGO 

The World's Fair at Chicago was noteworthy in the extensive use 
of non-theatrical movies for advertising and other purposes. 2 It is 
also certain that at other fairs of like world-wide importance, non- 
theatrical movies will play a particularly important part. Every 
conceivable kind of product was represented among the items ex- 
plained and sold by means of motion pictures automobiles, pocket 
flashlights, alloy metals, radios and radio batteries, welding gas, 
theater carbons, dentistry, scenery, transportation, travel, safety 
glass for automobiles, glass bottles for commercial and home canning, 
airplanes and airplane travel, headlamps for automobiles, refrigera- 
tors and electric stoves, tires, newspapers, mimeographing machines, 
toothpaste, gasoline, freight barge transportation, floor scrubbers, 
college education, fire extinguishers, novel double-keyboard pianos, 
marvels of curative medicine, the horrors of warfare, mine products, 
stockings, shoes, paints, carpets, agricultural machinery, tractors 
and trucks, canned foods of every kind, personal loan services, special 
educational methods for handicapped children, electric power appli- 
cations, bread and yeast, cod-liver oil, patent health products, steel, 
office machines, every type of social service and, of course, motion 
pictures to sell motion pictures. 

For example, two very noteworthy set-ups at the Fair will be 
mentioned. One was a model of an ordinary flashlight about twenty 
times the actual size. In conjunction with this, a continuous 16-mm. 
projector was aligned to project a picture upon a screen that took the 
place of the flashlight "lens." The interesting feature of the 
set-up was that the projector was started by speaking into a micro- 
phone. The impulse from the microphone started the projector so 
that it ran a complete cycle and then stopped. 

The other instance involved the use of an automobile operating on 
rollers. In front of the automobile was a screen on which were pro- 
jected pictures of other automobiles crossing and recrossing a busy 
intersection. The driver in the test automobile operated his car, 
changed gears, put on the brakes, as dictated by the driving conditions 
shown on the screen. The driver's trials came up thick and fast, 
requiring quick and correct thinking and acting. After the test the 
driver was given a certificate rating his ability. As may be appre- 
ciated, the test was watched with great interest by large audiences. 

At the close of the 1933 Fair questionnaires were circulated among 
the manufacturers. From these and other data a survey was made 



Jan., 1935] NON-THEATRICAL EQUIPMENT COMMITTEE 25 

from which the following facts were extracted. Seven users of ad- 
vertising motion pictures said they would use movies for 1934 and 
assign greater prominence to them; seven were undecided; eight 
would assign the same space; one would not use them; twenty-six 
out of thirty declared movies an important help, and fifteen of the 
twenty-six characterized them as of "primary importance." 

The non-theatrical film used at the Fair may be classified roughly as 
follows : 

(a) To stop and attract passers-by. 

(6) Institutional telling the story of the manufacturer and of how goods 
are made. 

(c) Showing products in use, to educate and sell. 

(d) Instructional training salesmen. 

(There are other uses of films such as training new factory help, micro motion 
analysis, machine design, etc.) 

Briefly, the advantages 3 of 16-mm. equipment and film are: low 
cost, no fire hazard, greater compactness, easier to operate, no ex- 
pensive operators required, continuous attachments more practicable, 
lower maintenance cost, and longer film life. 

With a few minor exceptions, 16-mm. film was used almost gen- 
erally at the Fair. The following figures cover the 16-mm. equipment 
used at the Fair during 1933 : 

16-mm. silent projectors 61 

sound-on-film projectors 19 

sound-on-disk " 11 



91 

Projectors manually operated 17 

automatic continuous 74 



91 

Constant operation 51 

Automatic cyclical control 14 

Intermittent schedule 16 

Audience control (by push-button) 10 

91 

Projectors placed out in open 18 

built into display features 17 

" " special cabinets 16 

concealed behind walls 34 

used from proj. booths 6 

91 



26 NON-THEATRICAL EQUIPMENT COMMITTEE [J. S. M. P. E. 

Front projection (32) Direct projection 60 

Rear projection (56) 1 Mirror 13 

2 " 16 

3 " 2 



91 

About Vs of the installations used screens 30" to 40" 
About Vi " " " " " 12" to 20" 

About V? " " " " " 20" to 30" 

Mostly 500-watt lamps were used; next in favor was the 400-watt lamp; 
a few 750-watt units were used. 

The final figures are not available for the 1934 Fair, but they are so 
close to those presented above that the data may be considered quite 
representative, though a greater number of 750-watt units have been 
employed. The utmost illumination possible is needed where screen 
brilliance has to compete with a relatively high level of general 
illumination. The consistent success of 16-mm. equipment under 
these most arduous conditions is most encouraging. 

REFERENCES 

1 Report of the Non-Theatrical Equipment Committee, /. Soc. Mot. Pict. 
Eng., XXIII (July, 1934), No. 1, p. 9. 

2 "Behind the Scenes at the Century of Progress World's Fair," Bell & Howell 
Co., Chicago, 111. 

3 DUBRAY, J. A., AND MITCHELL, R. F. : "A Parallel of Technical Values 
between 35-Mm. and 16-Mm. Films," Amer. Soc. Cinemat. Annual, 2 (1934), 
p. 329. 

A SHORT SELECTED BIBLIOGRAPHY COVERING THE USE OF MOTION PICTURES 
FOR EDUCATIONAL PURPOSES 

The use of motion pictures for educational purposes has been ap- 
preciated for a long time, but it has seemed to come rapidly to the 
fore within the past year or so. Several thousand schools throughout 
the country are equipped with 16-mm. projectors, both silent and 
sound, and quite a few universities and other special educational 
film exchanges are becoming more busily employed in the rigid and 
efficient exchange of educational subjects. Important international, 
U. S. Government departmental, and other authoritative investiga- 
tions being published indicate still more rapid growth in this par- 
ticular field. 

The following comments on some of the literature covering this 
field have been submitted by various members of the Committee, and 
are believed to be up-to-date, authoritative, and comprehensive. 



Jan., 1935] NON-THEATRICAL EQUIPMENT COMMITTEE 27 

Books: 

"Motion Pictures in Education in the United States" (University of Chicago 
Press); C. M. Koon, Senior Specialist in Radio and Visual Education, U. S. 
Office of Education. 
A complete and comprehensive survey compiled for the International Congress 

of Education and Instructional Cinematography at Rome in April, 1934, originally 

published as Circular No. 130, of the U. S. Department of the Interior, 1934. 

The chief headings are: 

(1) The Educational Influence of Motion Pictures. 

(2) The Motion Picture in the Service of Health and Social Hygiene. 

(3) The Motion Picture in Governmental Service. 

(4) The Use of Motion Pictures in Vocational Education. 

(5) The Motion Picture in International Understanding. 

(6) Motion Picture Legislation. 

(7) The Technic of Making and Displaying Motion Pictures. 

(8) The Systematic Introduction of Motion Pictures in Teaching. 

(9) Educational Problems of a General Nature Resulting from the Systematic 
Introduction of Motion Pictures in Teaching. 

(10) General Conclusions. 

"The Educational Talking Picture," F. L. Devereux, University of Chicago Press, 

1933. 

This book presents a general view of the field of the educational talking picture. 
Chapters XI and XII discuss school building requirements for audio-visual 
instruction, and types of equipment and standards for selecting them. These 
chapters include such topics as "Past Provisions Made in School Buildings for 
Visual Education Equipment"; "The Physical Provisions Required for Audio- 
visual Teaching"; "The Audio- Visual Studio Its Use, Its Location, and Its 
Desirable Characteristics in Various Types of School Organization"; "Detailed 
Standards for Planning and Equipping the Auditorium"; "Detailed Standards 
for Planning and Equipping the Music Unit"; "Provisions for Pupils with 
Hearing Difficulties"; "Acoustics in School Buildings"; "The Ventilating 
System"; "Relative Loudness"; and the like. Chapter XII presents the 
general standards for selecting equipment of various types. 

"Measuring the Effectiveness of Talking Pictures as Teaching Aids"; V. C. 

Arnspiger, Teachers College, Columbia University, New York, 1933. 

Special attention might be called to Chapter VI, in which a rather objective 
analysis of the effectiveness of certain technical elements of sound-film compo- 
sition is made. Attention is called also to the data regarding focal length of 
scenes; the quality of lighting in the films; speech, other sounds, and picture; 
repetition and integration of audio-visual elements. Attention might well be 
given to Chapter VIII, in which problems for future research are discussed. 
In this short chapter emphasis is placed upon the possibility of extending the 
range of the present curriculum by utilizing the sound motion picture as the 
most recent device for presenting subject matter. 

"Visual Instruction"; McClusky, et al., Motion Picture Producers & Distributors 
of America, Inc., New York, N. Y. 



28 NON-THEATRICAL EQUIPMENT COMMITTEE 

"Sound Pictures as a Factor in Education"; Fox Film Corp., New York, N. Y., 

1931. 
"Motion Pictures in the Classroom"; B. D. Wood and F. N. Freeman, Riverside 

Press, New York, N. Y. f 1929. 
"Visual Instruction in the Public Schools"; A. V. Dorris, Ginn & Co., New York, 

N. Y., 1928. 
"Motion Pictures for Different School Grades A Study of Screen Preferences"; 

M. A. Abbott, Teachers College, Columbia University, New York, N. Y., 1928. 
"Visual Fatigue of Motion Pictures"; A. Singer, Amusement Age Pub. Co., 

New York, N. Y., 1933. 
"Motion Pictures for Instruction"; A. P. Hollis, The Century Co., New York, 

N. Y., 1926. 

Periodicals: 

The Educational Screen (monthly) ; Chicago, 111. 

International Review of Educational Cinematography (monthly) ; Rome, Italy. 

As an example, the January, 1934, issue contains the following articles: 
"Problems Involved in the Development of Educational Talking Pictures"; 

N. L. Engelhardt, Teachers College, Columbia University, New York. 
"The Use of Taking Pictures in the Elementary School"; J. A. Brill, Erpi Picture 

Consultants, Inc., New York. 
"Sound Pictures as Factor in Class Size"; A. J. Stoddard, Superintendent of 

Schools, Providence, R. I. 
"Will Sound Pictures Remake the Curriculum?" P. R. Mort, Teachers College, 

Columbia University. 

(The International Education Institute is conducting an international sym- 
posium on the use of motion pictures for educational purposes. For two years 
they have been compiling the cinematographic encyclopedia, which they hope 
to complete soon.) 

R. F. MITCHELL, Chairman 

H. A. ANDERS W. B. COOK H. C. HOLSLAG 

V. C. ARNSPIGER H. T. COWLING A. SHAPIRO 

D. BEAN H. DEVRY C. TUTTLE 

E. W. BEGGS R. E. FARNHAM A. F. VICTOR 

H. GRIFFIN 



REPORT OF THE COLOR COMMITTEE* 

For some months past, the Color Committee has been engaged in 
preparing a glossary of technical and proprietary terms used in color 
cinematography. At first thought it may appear that such a contri- 
bution might be a waste of time and effort, and that the attention of 
the Committee might more profitably have been turned to something 
more creative. There are, however, certain factors that raise the 
labors of the Committee out of the category of the inconsequential. 

The various arts and sciences associated with the uses of color are at 
variance in their nomenclature, and in the language of the public at 
large discrepancies are even more marked. The very idea of color is 
so ill-defined in the minds of most persons that the inconsistencies of 
use of the term in ordinary conversation are sometimes absurd. We 
may imagine that, at the present moment, one of the lady guests of 
the Convention is asking a friend, "Are you wearing a colored gown 
to the banquet tonight?" The friend, thinking of her black dress or 
her white dress, answers, "No." But had she been asked, "What 
color dress are you wearing tonight?" the answer would have been 
"black" or "white" without hesitation. 

At the present time, other agencies are working to standardize the 
nomenclature of color. The Colorimetry Committee of the Optical 
Society of America prepared for the fall meeting at Washington a 
report modifying some of the definitions of its previous report of 1925. 
The U. S. Bureau of Standards likewise has recently put forward a 
proposal to change color nomenclature. It seems fitting, therefore, 
that the Society of Motion Picture Engineers, which is concerned 
with color both as a science and an art, should propose for general 
acceptance some of the special terminology of its trade. 

The Committee has, accordingly, compiled a glossary of some two 
hundred terms in common use in the industry. This glossary will be 
presented in the pages of a forthcoming issue of the JOURNAL. 

The list of terms treated by the Committee has been gathered from 
many sources: Our own JOURNAL, books on color photography, and 
the colorimetry reports of the Optical Society have been consulted, 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 

29 



30 COLOR COMMITTEE 

and supplementary lists have been received from various persons 
engaged in color cinematography. Some processes of color cine- 
matography that appear to be of no great commercial importance 
have been omitted for the sake of brevity. Technical terms for 
which satisfactory definitions are easily found in text-books of 
chemistry and optics are included in only a few instances. 

The last glossary published by the Society that of the Standards 
Committee in 1931 contained relatively few of the terms of our 
present effort. We wish, therefore, to call the attention of the 
Committee on Standards and Nomenclature to our listing, and to 
suggest that many of the terms contained therein should be incorpo- 
rated into any forthcoming glossary to be subjected to the processes 
of standardization. 

C. TUTTLE, Vice-Chairman 

P. D. BREWSTER R. M. EVANS J. F. KIENNINGER 

L. A. JONES 



REPORT OF THE HISTORICAL AND MUSEUM 
COMMITTEE* 

The work of the Historical Committee has progressed with good 
results. Besides many new accessions received for the collection at 
the Los Angeles Museum and the reconditioning of apparatus re- 
ceived, a number of biographies and autobiographies are in the course 
of preparation. 

These biographies will describe the experiences of the more notable 
pioneers in their achievement of the motion picture. Their successes 
and failures in groping toward the 'living picture" will be set down in 
an honest record of their accomplishments. So much has been 
written about the activities of the pioneers of the cinematograph from 
hearsay and memory, without foundation of fact, that it was thought 
desirable to create accurate accounts of the men who made the motion 
picture a possibility. The phenomenal growth of the art as well as 
the commercial aspects of the motion picture had done much to keep 
alive a great number of stories and traditions that credited the in- 
vention of many cinematographic devices to the wrong persons. In 
all fairness this should be righted while it is still possible to do so. 

Among the more valuable accessions to the collection is an accumu- 
lation of U. S. Patent Papers, brought together by the late Jean A. 
LeRoy and presented to the Society for the historical display at the 
Los Angeles Museum by Mrs. Jean A. LeRoy. Among the patent 
papers represented are those relating to motion picture devices from 
1860 to the present. The collection is an extensive representation of 
the patent literature, particularly of the years circa 1900. The last 
fifteen years are not so complete. A number of the patent papers are 
being placed on exhibition in swinging frames in the Motion Picture 
Gallery at the Los Angeles Museum. 

Another accession to the exhibit -is an Edison Exhibition Model 
Projector of the type introduced about 1901. It is complete with an 
arc light, double lenses for stereo projection, lantern slide arrange- 
ment and various devices used by a projectionist of the time. The 
projector was semi-portable and was contained in a wooden box so 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 

31 



32 HISTORICAL AND MUSEUM COMMITTEE [J. S. M. p. E. 

that the whole equipment could be carried about by the travelling 
showman of that period when the motion picture was shown in vacant 
stores as a vaudeville act, or as a feature of a travelling carnival, and 
at fairs. The showman, who was also, as a rule, the projectionist, 
carried the wooden box with the projector, a wooden base and iron 
legs, the lamp house with either arc light or lime-light, and the films 
and other incidentals. 

Harrison and Harrison, the filter manufacturers, presented a set of 
filters, which are being displayed so as to show the visual effects of the 
various filters. Besides the color-absorption filters, diffusing filters, 
filters used for producing night effects, and monochromatic viewing 
filters are shown. The filters will be displayed in front of a suitably 
colored lithograph so that comparative effects may be seen. 

C. R. Hanna, of the Westinghouse Electric & Manufacturing 
Company, presented a ground noise reduction device used in sound 
recording circuits, and Electrical Research Products, Inc., loaned a 
sound recorder and various slits, which have been placed on display. 

Among the more interesting loan items is a magic lantern with 
hand-painted slides of about one hundred years ago. The slides 
were panoramic views, and as the lantern used only a candle for 
illumination, only a small portion of each slide could be shown at a 
time as it was slid through the lantern. One interesting slide in the 
collection is a panoramic painting of the Tom Thumb , one of the first 
steam trains on the Baltimore and Ohio Line of 1829-30. This 
lantern was loaned by A. Falvy through the courtesy of Victor 
Merlo. 

Mrs. J. Trebaol loaned a praxinoscope, made by Emile Reynaud in 
1877. It consists of a drum mounted upon a stand for revolving. 
Instead of slits through which the pictures were viewed as in the 
"wheel of life," the device carried a series of mirrors arranged as 
facets at the center of the drum. When the drum was revolved, the 
various mirrors reflected the different progressive poses. The 
praxinoscope is believed to be the first device that showed progressive 
movement in its drawings. In the other hand-drawings and chrono- 
photographic devices, a horse, for example, would appear to be running 
upon a continuous belt in front of a fixed portion of the background, 
whereas in the Reynaud drawings the movement of the subjects was 
progressive. A man riding a horse, for instance, appeared to be going 
somewhere. Ransom Matthews, of the Los Angeles Museum, is 
making a model of the praxinoscope complete in every detail as com- 



Jan., 1935] HISTORICAL AND MUSEUM COMMITTEE 33 

pared with the original, so that if the original should be recalled from 
loan, an accurate model will be available. 

A large collection of hand-painted travel and astronomical slides 
of about 1825-35 has been located and purchased for the exhibition. 
They are beautifully painted, and each one is a masterpiece in crafts- 
manship and detail. They are circular and vary in size from three 
inches in diameter or less. Each slide glass is mounted in a wooden 
frame. Among the astronomical slides is one entitled A diagram that 
Proves the Rotundity of the World, which consists of a painting of the 
earth on one glass, and of a ship on a second glass. The picture of the 
ship is in contact with the picture of the earth, the path of the ship 
coinciding with the circumference of the earth. During projection, 
the ship is made to move around the earth by a gear arrangement, 
demonstrating the "rotundity of the world." 

A number of collections of photographs, posters, colored litho- 
graphs, autographed photographs, props used in making pictures, and 
apparatus have been presented. The collections are too numerous to 
mention, but their receipt has been appreciated, and they will form a 
valuable record for posterity. 

Persons having apparatus, literature, books, newspaper clippings, 
or other material portraying the history of the motion picture should 
communicate with the chairman of the Historical and Museum Com- 
mittee. It must be remembered that many valuable collections 
requiring years to bring together have been scattered or destroyed 
after the demise of, or loss of interest by, the collector. Such collec- 
tions must be preserved, and this Committee will assure the proper 
deposition of whatever is offered so that the material will be preserved 
for posterity. 

A record of the present as well as of the past is being preserved. 
A number of publishers of magazines and periodicals have placed the 
Museum on their mailing lists. Among them are The Motion Picture 
Daily, Hollywood Reporter, American Cinematographer, International 
Photographer, and others. 

Appreciation is extended to Sanford H. Place, who spent much 
time in reconditioning and overhauling historic apparatus at the Los 
Angeles Museum, Dick Rickard, of the Walt Disney Productions, 
Kenneth MacKaig, of United Artists, Miss Marcella Peterson, Fred 
Archer, Frances Christeson, Coleman Sellers, 3rd, Leon Gaumont, 
Louis Lumie"re, Wallace Clendenin, and many, many others. 



34 HISTORICAL AND MUSEUM COMMITTEE 

W. E. THEISEN, Chairman 

G. A. CHAMBERS W. V. D. KELLEY 

O. B. DEPUE G. E. MATTHEWS 

J. A. DUBRAY T. RAMSAYE 

DISCUSSION 

MR. CRABTREE: I hope that any of the members who have items suitable for 
the museum will not hesitate to send them to Mr. Theisen, because at last we 
have found a suitable repository for such historical material. For many years 
we have been talking about gathering historical equipment and made an attempt 
to put some in the Smithsonian Institute in Washington and in the Museum of 
Science and Industry in New York, but, unfortunately, we were not able to find 
some one in the East with sufficient interest in the work. Although I should 
like to see the establishment of a repository in the East, in the absence of some one 
who, like Mr. Theisen, is prepared to make this a labor of love, I think we should 
give our entire support to the Los Angeles Museum. 

MR. RICHARDSON: Some time ago I turned over to the Society a very valu- 
able Edison spool-bank projector in perfect condition, together with a reel of the 
original film and various other items. What has been done with them? 

MR. KALLMAN: We have the projector and film at the Museum of Science and 
Industry. We had planned to have a motion picture exhibit, but conditions at 
the museum prohibited our putting them on display immediately. They are in 
good order. 

MR. RICHARDSON: I believe, Mr. Chairman, we can't conduct two exhibits. 
The material should be sent to Los Angeles. 



REPORT OF THE PROJECTION PRACTICE COMMITTEE* 

Inadequate screen illumination has long been a problem of major 
importance in the projection field generally. Satisfactory screen illu- 
mination has been confined mainly to the larger theaters, the smaller 
theaters being either unable or unwilling to make the expenditure 
necessary to improve the quality of screen light. 

Exhibitors have recognized the need for more and better light on 
the screen; and they have recognized, too, the fact that the problem 
was more economic than technical. The situation, already serious, 
promised to become acute with the increasing use of color in motion 
pictures. 

Happily, the solution of the problem appears to be at hand, in the 
form of a new type d-c. projection arc which not only materially im- 
proves screen illumination but also satisfies the economic urgencies of 
the situation. This arc, using the new copper-coated Suprex carbons, 
is the topic of this report, the importance of which to the exhibition 
field is emphasized by the Committee. 

This report is based upon an extensive series of tests of the new arc 
made under actual operating conditions. Through the courtesy of 
various manufacturers there was made available to the Committee a 
group of motor-generators, both single- and three-phase rectifiers, 
and arc lamps of practically all the new types. 

The Committee desired to obtain the answers to the following ques- 
tions : 

(1) What is the carbon consumption per hour for values of current from 40 to 
50 amperes, using the 6- and 7-mm. combination; and from 50 to 65 amperes 
using the 6.5- and 8-mm. combination? 

(2) What is the ratio of burning of the positive and the negative carbons at 
different current densities? 

(3) What effect does the arc gap exert upon the burning rate of either carbon; 
and what arc gap affords the best results? 

(4) Is there a difference in arc voltage with different sources of supply, such as 
rectifiers and generators? 

(5) Is there any difference in the quality of the projected light when power is 
derived from either rectifiers or generators? 

(6) What increase of light occurs with an increase of carbon current density? 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 

35 



36 



PROJECTION PRACTICE COMMITTEE [J. S. M. p. E. 



(7) How do the various lamps now available compare as to light intensity, for a 
given current? 

(8) What is the efficiency and power-factor of the several power sources used 
with the new lamps? 

(9) What can be done to protect the reflector against pitting? 

(10) What are the over-all advantages of this new type of light source? 

RESULTS OF TESTS 

(1) Tests were made to determine first the rate of carbon consump- 
tion for various current densities, the arc gap being maintained con- 
stant that is, at from 6 /ie to n /32 inch; using as sources of cur- 
rent a polyphase rectifier, a single-phase rectifier, a motor-generator, 
and the regular d-c. power line. 




FIG. 1 



w j 

Rate of consumption of Suprex carbons; 6-mm. negative and 
7-mm. positive combination. 



The tests (Fig. 1) indicated that the consumption of the 6-mm. 
negative carbon was constant for currents between 40 and 50 amperes, 
being 3 3 /4 inches per hour or 5 / 8 inch for each ten minutes. The 
burning rate of the positive carbon, however, varied with the current. 
For 40 amperes, the rate was 6 3 / 4 inches per hour, or iy 8 inches for 
each ten minutes; for 45 amperes, lOVs inches per hour, or l n /i 
inches for each ten minutes; and for 50 amperes, 13V2 inches per 
hour, or 2 l / 4 inches for each ten minutes. 

Similar tests were made using the 6.5- and 8-mm. carbon trim at 
currents from 50 to 65 amperes, with the following results : 

For 50 to 57 amperes, the 6.5-mm. negative carbon burned at the 
rate of 3 8 /4 inches per hour, or 6 /s inch for each ten minutes which is 



Jan., 1935] 



PROJECTION PRACTICE COMMITTEE 



37 



identical to the burning rate of the 6-mm. negative at 40 to 50 am- 
peres. However, at 65 amperes, the 6.5-mm. negative carbon 
burned at the rate of 4 1 /2 inches per hour, or 3 / 4 inch for each ten 
minutes. 

The consumption of the 8-mm. positive carbon (Fig. 2) at 50 amperes 
was 6 inches per hour, or 15 /i 6 inch in ten minutes ; at 55 amperes, 8 x /4 
inches per hour, or ! 3 / 8 inches in ten minutes; at 60 amperes, 
10V2 inches per hour, or ! 3 / 4 inches in ten minutes; and at 65 am- 
peres, 13 x /2 inches per hour, or 2*/4 inches in ten minutes. 

(2) The burning ratio between the positive and negative carbons of 




66 



SO 52 54 56 S6 6O 62 4 t, 

FIG. 2. Rate of consumption of Suprex carbons; 6.5-mm. negative and 
8-mm. positive combination. 

the 6- and 7-mm. trim at 40 amperes is 1.8 to 1 ; at 45 amperes, 2.7 to 
1 ; and at 50 amperes, 3.6 to 1. 

With the 6.5- and 8-mm. trim the burning ratio at 50 amperes is 1.6 
to 1 ; at 55 amperes, 2.2 to 1 ; at 60 amperes, 2.6 to 1 ; and at 65 am- 
peres, 3 to 1. 

The tests definitely established that the burning time and the ratio 
of consumption of the positive and negative carbons do not change 
whether the arc current be supplied by a single-phase rectifier, a 
polyphase rectifier, or a motor-generator set, and that the design of 
the lamp also has no effect upon the aforementioned characteristics. 

The rate of consumption of the positive carbon is greatly affected 
by the current density; whereas the burning rate of the 6-mm. nega- 
tive carbon is affected little if any by the current density between the 



38 PROJECTION PRACTICE COMMITTEE [J. S. M. P. E. 

limits tested. However, the burning rate of the 6.5-mm. negative 
carbon varied slightly with the current density, particularly for cur- 
rents greater than 55 amperes. 

Although the operating limits of these carbon trims are generally 
understood to be 40 to 50 amperes for the 6- and 7-mm. trim, and 50 
to 65 amperes for the 6.5- and 8-mm. trim, the tests conducted by the 
Committee established definitely that best results are achieved when 
the trims are operated within the upper limits of their rated capacities. 

(3) Tests were conducted to determine the effect of the arc-gap 
upon the burning time and the burning ratio of the carbons. If the 
arc gap is increased, the current automatically decreases and the 
positive carbon then burns at a slightly lower rate. Likewise, if 
the arc gap be decreased, the arc current increases and the positive 
carbon burns faster. 

It is apparent, then, that since a change of current changes mate- 
rially the burning time of the positive carbon, and only slightly affects 
the burning rate of the negative, any change of arc gap will change the 
current and thus the ratio of burning of the positive and negative. 
This change of ratio tends to move the arc out of focus with the 
mirror. It is of the utmost importance, therefore, that the arc con- 
trol mechanism be sensitive enough to hold the arc-gap constant 
(Vie to n /32 inch), and that the current also be held constant if fre- 
quent focusing of the arc is to be avoided. 

Lamps having individual feed adjustments for positive and nega- 
tive carbons, thus allowing the burning ratio to be changed, permit 
adjustment to any desired current density within the limits of the 
rating of the carbons. However, lamps having single-feed screws are 
necessarily limited in operation to the current density, or burning 
ratio, for which the screw is designed. Deviation from the given ratio 
will entail constant attention and frequent manual adjustment. 

(4) In testing the 6- and 7-mm. carbon combination to determine 
what differences if any occurred in the voltage across the arc when 
various sources of power were used, single- and three-phase rectifiers 
and motor-generators were used. In each test the same lamp and 
the same carbons were used, and the same arc-gap was maintained 
under identical conditions. The results obtained are shown in Table 
I. It will be noted that there is a difference of 3 l / z to 6 volts across 
the arc in the case of the single-phase rectifier, and a difference of 
l /2 volt in the case of the three-phase rectifier, as compared with the 
motor-generator. This difference is due to the a-c. component of the 



Jan., 1935] PROJECTION PRACTICE COMMITTEE 39 

TABLE I 

Arc Voltages and Currents for 7 -Mm. Pos. and 6- Mm. Neg. Suprex Carbons 

(Source of Direct Current) (Amperes) (Volts) 

Three-PhaseRectifier 40 30 

Single-Phase Rectifier 40 27 

M-GSet 40 30.5 

Three-PhaseRectifier 45 32.5 

Single-Phase Rectifier 45 28 

M-G Set 45 33 

Three-Phase Rectifier 50 34.5 

Single-Phase Rectifier 50 29 

M-G Set 50 35 

rectified current. The d-c. voltmeter records only the d-c. value, and 
since the a-c. component is not registered, it is apparent that the 
greater the a-c. component, the greater will be the difference of volt- 
age as measured with a d-c. voltmeter. Similar tests were made 
with the 6.5- and 8-mm. carbon trim, with results as shown in 
Table II. 

TABLE II 

Arc Voltages and Currents for 8- Mm. Pos. and 6.5-Mm. Neg. Suprex Carbons 
(Source of Direct Current) (Amperes) (Volts) 

Three-PhaseRectifier 50 30.5 

Single-Phase Rectifier 50 27 

M-G Set 50 31 

Three-PhaseRectifier 55 32.5 

Single-Phase Rectifier 55 28 

M-G Set 55 33 

Three-PhaseRectifier 60 34.5 

Single-Phase Rectifier 60 29 

M-G Set 60 35 

Three-PhaseRectifier 65 38.5 

Single-Phase Rectifier 65 

M-G Set 65 39 

The tests indicate that the d-c. arc voltage for a given arc-gap de- 
pends upon the source of the current; but with the 6- and 7-mm. car- 
bons, the voltage will range between 30 volts at 40 amperes and 35 
volts at 50 amperes; and in the case of the 6.5- and 8-mm. combina- 
tion, between 30 volts at 50 amperes and 39 volts at 65 amperes the 
figures for both trims being based upon a current supply of acceptable 
smoothness. 



40 



PROJECTION PRACTICE COMMITTEE [J. S. M. p. E. 



(5) Single-phase rectifiers do not deliver current of the same 
smoothness as do three-phase rectifiers. The three-phase full-wave 
rectifier fills in with overlapping waves the gaps that exist when a 
single-phase rectifier is used (Fig. 3). Tests were made to deter- 
mine the visual effect of an alternating component upon the pro- 
jected light under normal operating conditions and with the shutter 
running at the standard speed of 90 feet a minute. With single-phase 
rectifiers the flicker was easily noticeable; whereas with both three- 
phase rectifiers and motor-generators there was no discernible flicker. 
These tests indicated that good screen results were not attainable 
with single-phase rectifiers. Both the three-phase rectifiers and the 
motor-generators delivered satisfactory results. 






A C COMPO*#r 



FIG. 3. Illustrating how the three-phase full-wave rectifier fills in the 
gaps in the current wave that exist when a single-phase rectifier is used. 

(6) To determine the change of light intensity for various values of 
current through the arc, the same optical system and the same measur- 
ing instruments were used throughout, and all tests were made with 
the shutter running. The results are therefore comparative, and are 
not computed in lumens per square foot (Fig. 4) . 

Burning the 6- and 7-mm. combination at 42 amperes, the arc con- 
sumed 1290 watts and the maximum average light intensity was 54 
units. At 45 amperes the arc consumed 1440 watts, and the light in- 
tensity was 65 units. .Thus, an increase of 11 per cent in wattage 
afforded an increase of 20 per cent in light. Burning the same com- 
bination at 50 amperes, the arc consumed 1700 watts, and themaxi- 



Jan., 1935J 



PROJECTION PRACTICE COMMITTEE 



41 



mum average light intensity was 80 units. Thus, an increase in watt- 
age of 32 per cent gave an increase of 48 per cent in light. This 
over-all increase was thus accomplished with a 19 per cent increase 
in current, or a 32 per cent increase in arc wattage (Fig. 5) . 

When the 6.5- and 8-mm. trim was burned at 50 amperes, the arc 
consumed 1540 watts, and the maximum average light intensity was 
70 units. At 55 amperes the arc consumed 1800 watts, and the 



70 



" 



F/G 



6-7 



8 



GHT 



100 



/20 



FIG. 4. Variation of light intensity with arc current. 



I 70 



1 

5 "' 



40 



/o oo 



6- 



CA/lt OH3 



Poi^fff /A 



OAn 



2000 
WAT\T3 CpHSdfffPBYAfa 



FIG. 5. Variation of power with light intensity. 

average light intensity was 84 units. Thus, an increase in arc 
wattage of 17 per cent provided an increase of 20 per cent in light. At 
60 amperes the arc consumed 2100 watts, and the average light in- 
tensity was 100 units. Here an increase in arc wattage of 36 per cent 
resulted in an increase of 43 per cent in light. At 65 amperes the 
arc consumed 2435 watts, and the average light intensity was 115 
units. Thus, with a total increase in wattage of 58 per cent the light 



42 PROJECTION PRACTICE COMMITTEE [J. S. M. P. E. 

was increased by 64 per cent. However, this represents a 30 per 
cent increase in the current. 

(7) Lamps of five different makes, designed for d-c. operation with 
Suprex carbons were tested and compared on the basis of projected 
light. The same projector was used in all the tests, only the lamps 
being changed. The results indicated that although the lamps all 
had reflectors of different sizes and focal lengths, the projected light 
in every case was of practically the same intensity for the same arc 
current. 

(8) In determining the efficiency and power-factor of the various 
sources of arc current, the following were considered : 

(A) 110- volt direct current from power mains. 

(B) Motor-generator, 80-volt d-c. output. 

(C) Motor-generator, 60-volt 

(D) Motor-generator, 40-volt 

(E) Motor-generator, double generator. 

(F) Single-phase rectifier. 

(G) Three-phase rectifier. 

Measurements were made of the over-all efficiency, or the propor- 
tion of direct current delivered to the arc with respect to the current 
drawn from the supply line (including the ballast resistance, in the 
case of the motor-generator), for a minimum load of 40 and a maxi- 
mum of 65 amperes. The values of efficiency follow: 

A (110-v., d-c. mains) 27 to 36 per cent. 

B (m-g., 80 v.) 26 to 35 per cent. 

C (m-g., 60 v.) 40 to 45 per cent, at 40 to 55 amperes. 

D (m-g., 40 v.) 45 to 48 per cent, at 40 to 50 amperes. 

(The limited capacities of the motor-generators in C and D did not permit test- 
ing them with the 60 to 65 ampere arc.) 
E (double generator, single-motor type) 45 per cent, at 45 to 50 amperes. 

(Not tested above 50 amperes, the rated capacity of the motor-generator.) 
F (single-phase rectifier) 48 to 55 per cent, at 40 to 50 amperes. 

(Capacity of rectifier, 50 amperes.) 
G (three-phase rectifier) 61 to 72 per cent, at 40 to 60 amperes. 

(Although the rated capacity of the rectifier was 60 amperes, the efficiency at 
65 amperes was 75 per cent.) 

The power-factor of the motor-generator sets tested ranged from 78 
to 83 per cent. The power-factor of the single-phase rectifier ranged 
from 80 to 85 per cent ; and of the three-phase rectifier from 85 to 90 
per cent. 



Jan., 1935] PROJECTION PRACTICE COMMITTEE 43 

(9) Examination of reflector mirrors in theaters in which Suprex 
carbons have been used for some time shows that there is continual 
pitting, resulting in a noticeable decrease in screen light. In order to 
maintain the screen illumination at its best, the mirrors should be 
replaced when noticeably pitted. 

There has been introduced recently a shield, or mirror guard, made 
with high-quality optical glass and having the same curvature as the 
mirror it is intended to protect. This guard fits exactly the inside 
curve of the mirror and acts effectively as a guard against pit- 
ting. Various sizes of mirror guards have been tested by the Com- 
mittee and found to occasion a negligible light loss. 

When the mirror guard itself becomes pitted, it can be easily re- 
moved and replaced with another, effecting a considerable saving over 
the cost of a new mirror. The Committee recommends the use of 
these guards. 

(10) The comparative advantages of the new d-c. light sources 
using copper-coated Suprex carbons may be judged on the basis of 
two factors: (a) quality and quantity of projected light, and (b) cost 
of operation, the latter being of primary importance to the smaller 
theaters. The following resume of operating cost per hour is based 
on the prevailing prices for carbons in standard shipping-case quan- 
tities, and an average current cost of 5 cents per kwh. : 

Carbon Cost per Hour 
(6- and 7-mm. trim, allowing for stubs) 

Amperes Cents 

40 10.4 

45 14.4 

50 18.5 

A re Current Cost per Hour 

(5 cents per kwh.) 

Amperes Cents 

40 6.0 

45 7.2 

50 8.75 

Supply Line Current Cost 

(Allowing for losses) 

D-C. Line 

and 80-V. Average 

Amperes Generator Generator Rectifier 

(cents) (cents') (cents') 

40 20 15 9 

45 24 17.5 11 

50 29 22 13.5 



44 PROJECTION PRACTICE COMMITTEE [J. S. M. P. E. 

Low-Intensity Costs, Cents per Hour, at 30 Amperes 
Carbon 4.9 

Current on line side of motor-generator, 80-volt type, at 5 cents 

perkwh. 19.7 

Cost with rectifier 11.5 

High-Low Costs, per Hour 

Carbons 17.2 

Current from line side of mot or -generator, at 5 cents per kwh. 47 

From the standpoint of quality and quantity of light, there is no 
reasonable basis for comparison between these new arcs using the 
Suprex carbons and low-intensity arcs, as there is a pronounced fa- 
vorable contrast in those respects in favor of the former. The low- 
intensity light is dull yellow; whereas the Suprex carbon arc delivers 
an intense white light which is very pleasing to the eye. A compari- 
son of the Suprex carbon arc with the high-low arc at a current of from 
50 to 60 amperes, showed that the Suprex carbon arc provides a light 
of equal intensity but with a more even field; and, of course, at a 
much lower operating cost. 

The Committee regards the d-c. Suprex carbon arc as one of the 
most important developments in the projection field within recent 
years. It fulfills the demand for improved screen illumination both 
as to quantity and quality of light in a manner that leaves no room 
for question. It enables the smaller theaters to offer for the first 
time a quality of screen illumination comparable with that found 
heretofore only in the largest and finest theaters. In addition, the 
Committee's test proved the arc to be economical in operation. 

Not only will the screen illumination benefit through use of this 
new type arc but the general illumination of the theater can be im- 
proved, certainly a very desirable advance. Colored motion pic- 
tures, the number of which is progressively increasing, demand a light- 
source of high intensity and good quality, a requirement that is 
fulfilled by this new arc. 

The Projection Practice Committee recommends the use of the 
Suprex d-c. arc. 

The Committee extends its thanks to the manufacturers who co- 
operated by supplying the equipment necessary for conducting the 
tests. The Committee is particularly indebted to the International 
Projector Corporation, which not only provided quarters in which to 
conduct the tests, covering a period of two weeks, but also contributed 
generously of its personnel, equipment, supplies, and electric power. 



Jan., 1935] PROJECTION PRACTICE COMMITTEE 45 

H. RUBIN, Chairman 

J. O. BAKER S. GLAUBER R. H. McCuixouGH 

T. C. BARROWS C. L. GREENE R. MIEHLING 

G. C. EDWARDS H. GRIFFIN P. A. McGuiRE 

J. K. ELDERKIN J. J. HOPKINS M. D. O'BRIEN 

J. J. FINN W. C. KUNZMANN F. H. RICHARDSON 

E. R. GEIB V. A. WELMAN 

DISCUSSION 

MR. RICHARDSON : I believe we should have included in the report the new 
a-c. light source, which is very economical and furnishes a very excellent projec- 
tion light. 

The statement was made that a glass guard causes negligible light loss. I 
understand that a loss of at least 4 per cent occurs for each polished surface of 
glass through which the light passes. You stated that the negative consumption 
curve rises gradually to a point after which, comparing it with the curve for 
the positive carbon, the negative burns rather steadily, indicating that the arc 
will remain at the focus of the mirror. But when the positive burns faster or 
slower than the negative, the light source will be out of focus constantly. In 
this particular case it would enlarge the spot. 

MR. RUBIN: As was stated hi the report, separate feed screws or adjustments 
for both negative and positive can adjust that. Actually this is a test curve. In 
the theater, the arc would be set for 50 amperes, and the adjustment would not 
be changed. 

MR. BRENKERT: As the current increases, the rate of consumption of the 
positive carbon increases faster than that of the negative. The procedure to 
follow as the current is increased is to step up the speed of the motor that feeds 
both carbons, then slow down the negative feed by a separate adjustment. On 
the arcs put out years ago that could not be done, but it can be done today on 
most arc lamps. 

MR. SACHTLEBEN: The light reflected by the light-guard will be 4 per cent 
at each surface, but because the surfaces of the light-guard are concentric with 
the surface of the mirror, the light reflected from these surfaces, except for second- 
order reflections and a very slight absorption in the glass itself, will be added 
to the light from the mirror. This loss of light will be very small, as was found 
by the investigations of the Committee. 

MR. BRENKERT: The best optical glass has a reflection loss at each surface, 
as Mr. Richardson stated, of 4 per cent. The guard provides two surfaces 
through which the light must pass on its way to the mirror, and on the return the 
light must pass through those surfaces again. Four such surfaces must be con- 
sidered. 

MR. RUBIN: In all tests that we made, with any instrument, the loss could 
not be detected. It can not be detected with the naked eye. That is why we 
could not report on it. The effect of using the guard is merely to make the mirror 
a little thicker. 

MR. BRENKERT: Is the guard made in the same mold as the mirror? The 
focal length of the reflector is an important factor, and it would, of course, be im- 
possible to put the guard hi exactly the same plane as that of the original reflector. 



46 PROJECTION PRACTICE COMMITTEE 

MR. RUBIN: That does not matter; you simply focus the combination of 
mirror and guard. The fact is the Committee discovered that the focus was 
just as good with or without the guard. The loss of light was negligible; we 
couldn't measure it. 

MR. SACHTLEBEN: These reflections do occur, but most of the light so re- 
flected adds to the light coming from the mirror. In view of the theory of ap- 
plication of the guard, the findings of the Projection Practice Committee are 
perfectly acceptable. 

MR. RICHARDSON: The Committee may have used a guard the surface of 
which happened to fit exactly. There are about fifteen different mirrors on the 
market and they are not all made on one tool. 

MR. BRENKERT: Elliptical reflectors are not ground and polished. In other 
words, they vary in focal length as well as in working distance, or both. It 
may be possible to make a few sufficiently accurately to superimpose one upon 
the other, but in production in large quantities I am afraid trouble will result. 
If you want satisfactory results, in practice all over the country and not in only 
a few spots, I should prefer to match up the protector with the reflector. Unless 
the problem can be put to the mirror manufacturers I believe it is entirely out 
of the hands of the lamp manufacturer to control the accuracy of the reflector. 
I want to be clearly understood: I don't object to anything that is an improve- 
ment, but I do object to anything that is going to cause trouble. 

PRESIDENT GOLDSMITH : It may be, of course, that extreme accuracy is not as 
necessary as we perhaps think. If the auxiliary guards are thin and if their 
surfaces are reasonably clean, then the major loss of light, of useful light, will 
result from absorption in the glass, rather than from reflection, and that neces- 
sarily will be fairly small in good, clear glasses. 

MR. RUBIN: The only question involved here is this: After a week or two of 
using the Suprex carbons, which pit considerably, will you get more light with 
the protector, which you claim causes a loss of 4 per cent, than from a pitted 
mirror? When you sell the exhibitor a mirror, he is not going to buy another one 
in two minutes. He will allow the mirror to become pitted more and more, for 
three weeks or a month, or perhaps a year. Which is to be preferred? 



THE NON-ROTATING HIGH-INTENSITY D-C. ARC 
FOR PROJECTION* 

D. B. JOY AND E. R. GEIB** 

Summary. The non-rotating high-intensity d-c. arc is considered in conjunction 
with equipment that has been developed for applying it to motion picture projection. 
The current and consumption ranges and the crater opening of carbons when used in 
this equipment are tabulated. The effect of burning the arc at various currents, arc 
lengths, and positions is shown graphically in relation to the light upon the projec- 
tion screen. The important features of the lamp mechanisms and optical systems 
are discussed. The sources of power used with this new type of arc are grouped in 
relation to arc stability and general characteristics. The influence of external mag- 
netic flux upon arc stability and appearance of the arc is also illustrated. 

INTRODUCTION 

The non-rotating high-intensity d-c. arc has been described in an 
earlier paper. 1 This arc has the characteristic cup-shaped crater of 
the well-known d-c. high-intensity arc with the rotating positive 
carbon, but is produced with smaller size carbons. It also has the 
desirable features of requiring no rotation of the positive carbon and 
of eliminating the necessity for conducting the current into the carbon 
close to the arc. 

It was believed at the time this arc was first announced that, with 
a suitable lamp mechanism and power supply, it could be used to ad- 
vantage for motion picture projection in houses in which more light 
and a better quality of light were desired than could be furnished by 
the low-intensity lamp, but in which the more complicated and ex- 
pensive apparatus for burning the conventional high-intensity d-c. 
arc was neither required nor economically justified. The need for 
this light source is more pronounced today than ever because of 
the great advances that have recently been made in the technic of 
producing colored pictures. The projection of colored pictures re- 
quires a higher intensity of light at the film aperture than the pro- 
jection of black-and-white pictures, and preferably the snow-white 
color characteristic of the light from the high-intensity arc. 

* Presented at the Fall, 1924, Meeting at New York, N. Y. 
** National Carbon Co., Cleveland, Ohio. 

47 



48 



D. B. JOY AND E. R. GEIB 



[J. S. M. P. E. 



In the previous paper the discussion was restricted to the arc and 
carbons themselves because, at that time, commercial equipment 
in which to use these carbons was not generally available. In this 
paper the arc is discussed in relation to equipment now available 
and to the light on the projection screen. 

DATA ON CARBON TRIMS 

The consumption and current-carrying capacity of the carbons 
has now been more thoroughly determined for equipment actually 
used in the theaters today, and the results are given in Table I. 
The values of current-can ying capacity and consumption differ 
slightly from those previously given because of the development of 
certain types of equipment that enables the carbons to maintain a 
steady arc at a lower current than was previously thought possible. 
These carbon trims have been designed for approximately the same 
ratio of consumption between positive and negative carbons for cor- 
responding burning rates. 

TABLE I 

Consumption and Current- Carry ing Capacity of Suprex Carbons 



Carbon 


Range of 


Consumption 
(Inches per Hr.) 


Effective Pos. 
Crater Diam. 
(Inches) 


Pos. Neg. 


Current 


Voltage 


Pos. 


Neg. 


6-mrn. 5-mm. 
7-mm. 6-mm. 
8-rnm. 6. 5-mm , 


32-40 
42-50 
56-65 


31-40 
31^0 
31-40 


6.5-13.5 
6.5-13.5 
6.5-13.5 


3.0-4.5 
3.0-4.5 
3.0-4.5 


0.14-0.16 

0.15-0.20 
0.20-0.25 



The effective crater diameters of the carbons are also given in this 
table. These figures were obtained by measuring the diameter of the 
inside of the crater edges, and are approximately 0.02 to 0.04 inch 
smaller than if the outside of the crater edge had been taken. This 
dimension is used so that the figures can be applied directly to 
calculating the magnification of the optical system of the projection 
lamp. 

EFFECT UPON THE LIGHT ON THE PROJECTION SCREEN OF CHANGING THE 
CURRENT AND POSITION OF THE ARC 

At the present time, most of the lamps and auxiliary equipment 
have been designed to use the 7-mm. positive and 6-mrn. negative 
carbon trim. We have, therefore, based our discussion of the light 
on the projection screen on this combination burned in a typical 



Jan., 1935] NON-ROTATING HiGH-lNTENSITY D-C. ARC 



49 



lamp. In accordance with previous experience, the light on the 
projection screen has been studied with a view of giving to the pro- 
jectionist, as well as the lamp and equipment designer, data that will 
enable him to obtain the best light possible from his particular piece 
of equipment. In actual practice there are several variables that 
simultaneously affect the light on the projection screen, but in this 
investigation the effect of each variable has been studied indepen- 
dently. By combining the data obtained on the several variables 
the resultant effect can be predicted. The average illumination of 







Avri 


MM SCWE 


1 U6HT 








































LOCATION 

USED 


Of POtriTS ON SCRtEM 

FOR LISHT RATIO 

JL 












A 


\ 




si 
















>il 
i 






UfiHT DlSTJ 


laurion 


















Jig 

.3 6 
.2 00 

-. f * 


















. 


, 



ARC LEMSTM IN INCHES 

FIG. 1. Light on projection screen vs. arc length: 
7-mm. pos., 6-mm. neg. carbons; positive carbon 3.76 
inches from reflector; constant current, 45 amperes. 

the projection screen was measured at nine points on the screen. 
The light distribution is expressed as the ratio of the light at the 
center of the screen to the light at the sides of the screen, as indi- 
cated by the diagram in Fig. 1. 

In ordinary practice the arc length of this 7-mm. positive, 6-mm. 
negative trim is usually maintained between 9 /32 and 6 /ie inch. The 
effect upon the screen light of varying the arc length, while keeping 
the current constant and the positive crater at exactly the same posi- 
tion with respect to the reflector, is shown in Fig. 1. It is obvious 



50 



D. B. JOY AND E. R. GEIB 



[J. S. M. P. E. 



ositior 
45am 



3^0 3ji 

AMCE IM IMCHES 
E CABBOM To Rt 



J.IMO xvuiiewv "i waabDS 





PH W> 

IS 

3 3 

'5* 5 bO 
&bjj 

ill 



O* ^ m 

s S 

Nl 









? ^g-s 

ii? 

s .^w,^ 

ff O D C 

a c 2 



^L o> 



Jan., 1935] NON-ROTATING HiGH-lNTENSITY D-C. ARC 51 

from these curves that neither the total light on the screen nor the 
distribution of the light is materially affected by changing the arc 
length from 3 /i 6 to 3 / 8 inch, provided the current and the position of 
the positive carbon remains constant. However, with a very short 
arc length, such as 3 /ie inch, there is built up on the negative carbon 
a reddish deposit which, unless removed, may cause difficulty in 
re-striking the arc. If, on the other hand, the arc length is com- 
paratively great, say, 3 /g inch or more, there is a perceptible wavering 
of the arc which tends to cause a fluctuation of the screen light. 

If the current is increased but the arc length and the position of 
the arc with respect to the mirror are held constant, there is a very 
definite increase in the screen light but very little change in light 
distribution, as illustrated in Fig. 2. For an increase in current 
from 40 amperes to 50 amperes, or 25 per cent, the light on the 
screen is increased by 47 per cent. This increase in light is accom- 
panied by an increase in crater depth and carbon consumption. If 
the arc current is too small, the crater is very shallow and the light 
is not uniform in color. If the current is too great, the consumption 
is excessive and the light is unsteady. 

If the current and arc length are maintained constant but the arc 
itself is moved with respect to the reflector, the screen light and dis- 
tribution vary as indicated in Fig. 3. The shape of the curve for 
average screen light is similar to that obtained for the 8-mm. high- 
intensity a-c. arc or the 12-mm. low-intensity d-c. arc with the same 
6:1 magnification of the optical system, 2 - 3 but has a much sharper 
peak and the light distribution ratio drops off more rapidly with in- 
creased distance from crater to reflector. In other words, to main- 
tain a good distribution of light upon the screen, it is necessary to 
hold the position of the positive crater within close limits. 

At A in Fig. 3 the light at the sides of the screen is equal to that 
at the center of the screen and the edge of the positive crater is 
3.70 inches from the center of the reflector. If the arc is moved 
closer to the reflector the sides of the screen become brighter than 
the center, which is a very undesirable condition. If the arc is 
moved away from the reflector, the average light on the screen in- 
creases, and the light at the center becomes brighter than that at the 
sides until the point B on the curve is attained, at which the average 
light is at its maximum. Continuing the movement of the arc away 
from the reflector results in a decrease in the average screen light 
and an increase in contrast between the center and the sides. After 



52 D. B. JOY AND E. R. GEIB [J. S. M. P. E. 

point C is reached, the contrast between side and center becomes 
very noticeable. If the distance between points A and C is taken 
as an arbitrary range, this allows a movement of the arc over a 
distance of only 0.14 inch. It is therefore quite essential that the 
position of the arc be accurately maintained near the center of these 
limits. The narrow range is due to the combination of the curve of 
light distribution across the crater face, the magnification of the 
optical system, the diameter of the crater opening, and the dimensions 
of the film aperture. The dimensions of the film aperture are stand- 
ard; and the shape of the light distribution curve, with the peak in 
the center, is characteristic of d-c. high-intensity arcs. This leaves 
the magnification of the optical system and the crater opening as the 
controllable factors affecting the screen light. The crater opening 
is increased when a larger size positive carbon at a higher current is 
used, as indicated in Table I. It has been found by measurement 
that the 8-mm. positive carbon with the 6.5-mm. negative carbon 
does afford greater latitude of movement from the optimal position 
of the arc as well as more light. With this 8-mm. carbon burning 
at 60 amperes, the permissible movement of the arc, for the same 
relative change of volume and distribution of light, is increased ap- 
proximately 35 per cent. The other option would be to increase the 
magnification of the optical system. The use of either a larger 
carbon or an optical system of higher magnification would make it 
much easier to maintain a uniform light upon the screen. The arc 
will not operate satisfactorily unless the correct alignment of the 
positive and negative carbons is maintained. 

From the foregoing facts it is obviously essential that the carbons 
feed uniformly, so as to maintain a given position of the positive 
carbon and a steady current. This is very effectively accomplished 
in a number of the lamps now on the market. The positive carbon is 
fed continuously, and the negative either continuously or at very 
short intervals of time, giving practically the same results. The 
ratio of the feed of the positive and negative carbons can be adjusted 
either in the lamp itself, or by a comparatively simple change in 
the lamp parts. If it is desired to change the feed ratio without 
adjusting the feed mechanism, it should be remembered that, for 
any given carbon trim, as the current is increased the ratio of the 
positive consumption to the negative consumption increases. 



Jan., 1935] NON-ROTATING HiGH-lNTENSITY D-C. ARC 
THE POWER SOURCE AND ARC STABILITY 



53 



All the preceding discussion assumes the maintenance of a stable 
arc, but it has been found that the stability of this type of arc is very 
materially affected by the source of power and the magnetic flux. 

The power source that is commonly used, whenever available, is 
the 115- volt d-c. line. This gives an opportunity to use plenty of 



I 

I" 



30 



\ 




B4C 




\ 



CURREMT AT AKC 

FIG. 4. Volt-ampere characteristics of 
arc due to power source: (.4) Constant- 
voltage source, 115 volts; resistor set for 
45 amperes, 35 volts at arc. (B) Con- 
stant-voltage source, 45 volts ; resistor set 
for 45 amperes, 35 volts at arc. (C) Vari- 
able-voltage source with falling volt-ampere 
characteristic. (D) Variable-voltage source 
with rising volt-ampere characteristic. 

ballast resistance between the arc and the line, which is necessary, 
with the low-intensity arc, for stable operation. However, it has been 
found that the low-intensity arc (e. g., 30 amperes and 55 volts on 
12-mm., 8-mm. S.R.A. carbons) is stable with 80 volts on the line. 
Therefore, in places where 115-volt direct current is not available, a 



54 D. B. JOY AND E. R. GEIB [J. S. M. P. E. 

motor-generator set has been used to provide d-c. power service of 80 
volts. 

These small size d-c. high-intensity carbons require only about 
35 volts at the arc. It would therefore be very uneconomical to use 
either a 115-volt or an 80- volt d-c. source of power. This fact has 
led to a consideration of sources of direct current, including both 
generators and rectifiers, that would be more economical than the 
80- volt motor-generator sets or 115-volt d-c. line. Several such 
devices have been developed which give very good results, with a 
noticeably steadier light on the screen than when the high-voltage 
source, such as the 115-volt d-c. line, is used. 

A consideration of the volt-ampere curves at the arc with these dif- 
ferent types of power sources, together with some of the characteristics 
of the high-intensity arc, will explain the reason for this seeming re- 
versal of well-known facts. The power sources considered had the 
following characteristics : 

(A) A constant voltage of 115. 

(B) A constant voltage of 45. 

(C) A falling volt-ampere curve and comparatively low no-load 
voltage. 

(D) A rising volt-ampere curve. 

With power sources A and B, which have constant-voltage char- 
acteristics at the currents that will be considered, a resistance is 
placed between the source and the arc to reduce the voltage to the 
proper value. With a power source such as C or D, the character- 
istic of the source itself is such that voltage of the proper value is 
supplied at the operating current. In considering the stability of 
the arc we are particularly interested in the volt-ampere curves at 
the arc itself. The effect of the various power sources on the volt- 
ampere curve at the arc through a small range of current is shown 
in Fig. 4 for a 45-ampere, 35-volt arc with 7-mm. positive and 
6-mm. negative carbons. 

Bassett 4 has shown that the voltage drop across the arc stream 
from the positive to the negative is comparatively low and does not 
increase materially as the current is increased. On the other hand, 
the voltage drop in the positive crater is comparatively high and 
increases materially as the current and crater depth increase. As 
shown in Fig. 2, this increase of voltage and crater depth at the posi- 
tive carbon that accompanies the increase of current, causes a very 
material gain in the useful light at the arc. It is therefore very im- 



Jan., 1935] NON-ROTATING HlGH-lNTENSIY D-C. ARC 



55 



portant that conditions at the positive crater be maintained as con- 
stant as possible. If for some reason there is a disturbance in the 
high-intensity effect, such as, for example, a decrease in the voltage 
drop in the crater caused by a crooked crater due to poor alignment 
of the carbons, the resultant effect on the arc may be quite different, 
depending upon the characteristics of the power source. 

The curves in Fig. 4 show only the influence of the power source 
on the effect that arc disturbances have upon the arc current and 
voltage. They do not take into consideration changes in the char- 
acteristics of the arc itself, which might in turn affect the volt- 
ampere characteristic. 




POSITIVE. CARBON 



NIG,* TIVC 

FIG. 5. Illustrating the distribution of the magnetic 
flux about the carbons set at an angle less than 180 de- 
grees. 

Over the small range of current and voltage that is being con- 
sidered, power sources B and C can be regarded as having approxi- 
mately the same volt-ampere curve. With power source A, where 
the voltage is cut down from 115 volts to 35 volts by a comparatively 
large amount of ballast resistance, the volt-ampere curve is very 
steep. A decrease of arc voltage would cause a small increase of 
current from the power source. Power sources such as B and C 
display, for the same decrease of arc voltage, a much greater in- 
crease of current. On the other hand, power sources having a 



56 



D. B. JOY AND E. R. GEIB 



[J. S. M. P. E. 



rising volt-ampere characteristic would show a decrease of current 
with a decrease of arc voltage, as in curve D. It will be demon- 
strated that these differences in the action of the various sources of 
power on the arc materially affect the arc stability. 

If, for example, some disturbance causes a momentary decrease 
of the arc voltage of one volt, the immediate action of the power 
sources on the arc, as indicated by Fig. 4, is as shown in Table II. 

TABLE n 

Effect of Decrease of Arc Voltage of 1 Volt 
(Original Conditions 45 Amperes and 35 Volts at the Arc) 



Power Source 

(A) 115-volt constant- voltage power line 

(B) 45-volt constant-voltage generator 

(C) Generator or rectifier with falling volt-ampere 
curve similar to source B at and near 45 am- 
peres 

(D) Generator with rising volt-ampere character- 
istic 



Momentary 

Arc 
Voltage 

34 
34 



34 



34 



Momentary 

Arc 
Current 

45.6 
49.5 



49.5 



44.4 



With a slight increase of current in the arc, such as occurs with 
power source A, there would be very little tendency for the crater 

depth to be restored. On the other 
hand, if the current decreased as 
the arc voltage decreased, as in case 
D, the crater depth would be further 
diminished and the condition aggra- 
vated. But if the power source B 
or C were employed, there would 
be a distinct increase of current, 
which would immediately tend to 
restore the proper crater depth. 
This, in turn, would increase the 
arc voltage and cause a restoration 
of current, arc voltage, and crater 
depth to their normal values. It is 
therefore apparent that the use of 
a power source such as D would be 

intolerable, and the use of a power source such as B or C would be 
considerably better than power source A in maintaining an arc of 
constant characteristics. Power source D corresponds to the series 




FIG. 6. Photograph of arc under 
conditions illustrated in Fig. 5. 



Jan., 1935] NON-ROTATING HiGH-lNTENSITY D-C. ARC 



57 



arc generator operated below the knee of the current-voltage curve 
and, as stated by Dash, 5 is not suitable for even the conventional 
type of high-intensity arc with rotating positive carbons. The 80- 
volt generators of the constant- voltage type, common in theaters 
in which alternating current must be converted to direct current 
for low-intensity mirror arc projection, are, of course, intermediate 
in effect between power source A and power source B or C. Power 
sources B and C are typical of the power units that have recently been 
developed for use with the d-c. high-intensity arc using the small 
size non-rotating positive carbons. Several different types of such 
units are here listed: 

(1) A low-voltage motor-generator set with constant-voltage characteristic, 
and with the two lamps and their individual ballast resistances connected to the 
generator in parallel. 




FIG. 7. Illustrating uniformity of distribution of magnetic flux 
when carbons are aligned, without additional magnetic field. 

(2) A motor-generator set with a falling volt-ampere characteristic, which, 
because of its characteristic requires no ballast resistance, but does require a 
separate generator, possibly driven from a common motor, for each lamp. 

(5) A rectifier with a falling volt-ampere characteristic. This also requires 
a separate rectifying unit for each lamp. 

These three types of power sources are being manufactured and 
marketed for use with these non-rotating high-intensity d-c. arcs at 
the present time. Each power source has its own particular ad- 
vantages. All, however, are characterized by a comparatively 
large increase of the current with a decrease of arc voltage at the 
operating current and voltage of the arc. This is a decided advantage 
in maintaining stability of the arc. All these power sources are 
much more efficient than 80- or 115-volt constant- voltage generators 
with the necessary ballast resistances. It is very desirable that 



58 D. B. JOY AND E. R. GEIB [J. S. M. P. E. 

the short-circuit current of these sources of power be as small as 
possible, for, if the short-circuit current is excessive when the arc is 
struck, the crater is shattered and the core of the carbon tends to 
be blown out. This causes unsatisfactory light on the projection 
screen until a normal crater has been reestablished. 

If three-phase or two-phase rectification is used, the variations 
in the instantaneous values of the current, voltage, and light upon 
the screen are comparatively small and are not noticeable. If a 
single-phase rectifier is used, similar to those commonly employed 
for low-intensity mirror arc lamps, there is a considerable fluctua- 
tion of voltage and current which, with the high intensities of light 




FIG. 8. Photograph of arc under conditions illus- 
trated in Fig. 7. 

obtained with the high-intensity d-c. arc, might cause a noticeable 
beat of light upon the screen under certain conditions. If, however, 
a sufficiently large choke-coil is used on the output side of the single- 
phase rectifier, this fluctuation is materially decreased. 

THE USE OF MAGNETIC FLUX TO STABILIZE THE ARC 

Another very important factor to be considered in stabilizing 
this high-intensity arc is the use of an auxiliary magnetic flux. The 
ordinary high-intensity arc, in which the positive carbon is rotated 
and the current fed into the carbon close to the positive crater, has 
always been burned with the negative carbon set at an angle of 
more than 90 but less than 180 degrees to the positive carbon. 



Jan., 1935] NON-ROTATING HiGH-lNTENSITY D-C. ARC 



59 



Fig. 5 illustrates the distribution of the lines of force generated about 
the arc by the electric current. Because the negative is at an angle 
less than 180 degrees to the positive carbon, the lines of force are 
crowded together directly beneath the arc, while the flux density is 
less above the arc. This causes a resultant force in the direction 
indicated by the arrow, which, in conjunction with the natural flow 
of the arc stream, projects the tail-flame of the arc in an upward and 
forward direction from the positive crater, as shown in the photo- 
graph in Fig. 6. 

The lamps developed for burning the new small size, high-intensity, 




FIG. 9. Illustrating distribution of magnetic flux, using a supple- 
mentary magnetic flux. 

non-rotating positive carbons are of altogether different design from 
that of the older types of high-intensity lamps, and, since there is no 
necessity for rotating the carbon or feeding the current into the 
carbon close to the arc, the carbons are held and aligned just as in 
the low-intensity mirror arc lamps. In other words, the angle of 
the negative carbon has been increased until the negative and posi- 
tive carbons are directly in line. The lines of force generated by 
the arc current, as shown in Fig. 7, are distributed uniformly about 
the carbons, and there is no resultant force in the upward direction 
except that caused by natural drafts in the lamp. The tail-flame, 



60 D. B. JOY AND E. R. GEIB [J. S. M. P. E. 

therefore, surrounds the arc in almost a uniform layer, as shown by 
the photograph in Fig. 8. Unless the lamp design is such as to 
eliminate extraneous drafts or other disturbing conditions, a slight 
misalignment of the carbons causes the crater to burn off to one side. 
This in turn, particularly with a power source such as a 11 5- volt 
d-c. line, causes a noticeable change in the depth of the crater and 
the light upon the screen. If, however, a supplementary magnetic 
flux is provided to increase the flux density below the arc and de- 
crease it above, an upward force is established which materially 
changes the shape and characteristics of the arc. This magnetic 
flux has been introduced in some of the lamps by means of a magnet, 




FIG. 10. Photograph of arc employing the 
supplementary magnetic field; negative carbon 
lowered slightly. 



as in Fig. 9, which illustrates diagrammatically the carbons and lines 
of magnetic force as seen from above and in front of the positive 
crater. The flux generated by the current in the carbons surrounds 
the carbons uniformly, as in Fig. 7, but superimposed upon it are 
the lines of force emanating from the magnet. This flux from the 
magnet is in such a direction as to tend to neutralize the flux above 
the arc generated by the arc current, and increase the total flux be- 
low the arc. The tail-flame from the arc lengthens and is driven 
upward, becoming comparatively stationary and constant in length 
and direction. The negative carbon should be lowered slightly to 
compensate for this direction of the arc stream. There is then no 



Jan., 1935] NON-ROTATING HiGH-lNTENSITY D-C. ARC 61 

tendency for the crater to break away in any direction. A side view 
of the arc employing the auxiliary magnetic flux is shown in Fig. 10. 
The improvement attained in the steadiness of the light upon the 
projection screen by using the auxiliary magnetic flux is very notice- 
able when a 115-volt d-c. line is the power source. It is less notice- 
able with power sources such as B and C, because the high-intensity 
effect is more stable with power sources of those types. 

The effect of the type of power source and of the auxiliary mag- 
netic flux upon both the arc and the screen light are of very great 
practical importance, and it can be said without reservation that 
successful operation of this type of arc depends largely upon these 
two factors. This discussion emphasizes the fact that coordination 
of the light source, optical system, lamp mechanism, power source, 
and projectionist is essential to good light on the projection screen. 

REFERENCES 

1 JOY, D. B., AND DOWNES, A. C.: "Direct-Current Hign-Intensity Arc with 
Non-Rotating Positive Carbons," J. Soc. Mot. Pict. Eng., XXII (Jan., 1934), No. 
l,p.42. 

2 JOY, D. B., AND DOWNES, A. C. : "Properties of Low-Intensity Reflecting Arc 
Projector Carbons," /. Soc. Mot. Pict. Eng., XVI (June, 1931), No. 6, p. 684. 

3 JOY, D. B., AND GEIB, E. R.: "The Relation of the High-Intensity A-C. Arc 
to the Light on the Projection Screen," J. Soc. Mot. Pict. Eng., XXIII (July, 
1934), No. 1, p. 35. 

4 BASSETT, P. R.: "The Electrochemistry of the High-Intensity Arc," Trans. 
Amer. Electrochem. Soc. XLIV (1923), p. 153. 

6 DASH, C. C.: "Operation of Projection Arcs from Motor-Generator Sets," 
/. Soc. Mot. Pict. Eng., XV (Nov., 1930), No. 5, p. 702. 

DISCUSSION 

MR. PALMER: When an arc is operated by a motor-generator set, it pro- 
duces a perceptible hum. It probably isn't serious hi the projection room of a 
theater, but it is very serious when arcs are used in sound picture studios, where 
sound pictures are being made. Forty or fifty arcs make a lot of noise. 

Various means have been tried to stop the noise, and it can be stopped by 
putting choke-coils in the circuits ; but that is quite expensive. I wonder whether 
there is any possibility that it might be stopped by some magnetic scheme. In 
the first place, what is it that causes the sound, and is there any possibility of re- 
moving it by a magnetic means? 

MR. JOY: I have always believed that the sound was due to the "commutator 
ripple," which, of course, is a function of the number of commutator bars, poles, 
and speed of the motor-generator set. I don't believe that any external magnetic 
effect would decrease the hum. 



62 D. B. JOY AND E. R. GEIB 

The hum can be eliminated by a condenser across the line, a choke coil in the 
line, or, preferably, by a combination of condenser and choke coil. 

MR. PALMER: But what is it that actually causes the sound? 

PRESIDENT GOLDSMITH: Would it not be thermal expansion and contraction 
resulting from the changes of current causing changes in heating of the surround- 
ing air? 

MR. PALMER: I notice that the negative carbon can be adjusted until the 
positive crater is in a vertical plane, so as to keep it as nearly in the focus of the 
mirror as possible. 

MR. JOY: Yes. Of course, the external magnetic effect, as we demonstrated, 
tends to force the tail-flame upward. If the negative carbon were centered with 
respect to the positive carbon, the crater of the positive carbon would be shallow 
at the top. If the negative carbon is lowered a few thousandths of an inch this 
compensates for the upward sweep of the tail-flame and results in an even crater 
on the positive carbon. All the lamps using Suprex carbons have means for con- 
veniently adjusting the negative carbon. In reference to mirror pitting, it is 
our opinion that mirror pitting is not caused by particles of copper. The copper 
plate is melted away from the carbon by the heat of the arc for an appreciable 
distance back from the ends of the carbons and therefore does not get into the 
arc itself. The pits on the mirror are caused principally by particles of carbon 
or core driven out of 'the arc and striking the mirror when the arc is started and 
the momentary short-circuit current is very high. The particles are shot out from 
a reddish deposit on the negative carbon which is formed, as stated in the paper, 
when the arc length is too short; or occasionally they come from the arc itself. 
Some of the deposit on the mirror is red in color. This probably led to the belief 
that it is copper. 

MR. SAMUELS: In a certain chain of theaters using other types of power sup- 
ply, there was considerable pitting of mirrors. After using the Unitwin motor- 
generator which has a very low striking voltage and small short-circuit current, 
the pitting was greatly reduced if not almost entirely eliminated. We are not 
so much concerned as to the actual material that causes the pitting as we are 
in eliminating it. 

MR. ELDERKIN: Does not the pitting continue all the while the arc is burning? 
If it occurs only during the strike, an inside douser could prevent that. It is 
my impression that pitting goes on all the time the arc is burning. 

MR. JOY: It is our opinion that a very large proportion of the pitting occurs 
when striking the arc. Some of the new lamps, designed for these Suprex carbons, 
did not have an inside douser that fully protected the mirror. I believe that is 
largely corrected at the present time because the lamp manufacturers have come 
more and more to realize that it is essential, particularly with the high short- 
circuit starting current, that the mirror should be protected when the arc is 
first started. 



THE DEVELOPMENT OF 16-MM. SOUND MOTION 
PICTURES* 

E. W. KELLOGG** 



Summary. The history of 16-mm. development and some of the problems with 
which the engineers were confronted in the effort to obtain good sound quality are 
outlined. In view of the difficulties with 35-mm. film, and the greatly reduced speed 
at which 16-mm. film must run, the prospect of obtaining really satisfactory sound 
was at first far from encouraging; but with careful technic, commercial quality was 
obtained, and continuous progress has been made since. Elimination of printing 
losses by recording directly on each 16-mm. film from a good 35-mm. master print 
with suitable electrical compensation for high-frequency losses, gave the best films 
obtainable. Improvements in optical reduction sound printers have more recently 
made this system look extremely promising. Extended studies of film characteristics 
and processing, and improvements in optical recording systems have contributed in an 
important way to securing better sound. 

Projector design has centered about questions of compactness, convenience, and 
simplicity as well as performance. Several of these items are discussed, and ex- 
pedients employed are described. 

The advent of real amateur talking pictures calls for complete sound and picture 
recording apparatus of utmost portability and simplicity. In this undertaking the 
quality problems are renewed in more difficult form, and many new problems of 
electrical, optical, and mechanical, as well as photographic nature, had to be solved. 

During the early stages of the commercial development of "talking 
movies" for theaters, with the sound recorded photographically on 
the edge of the film, the problem of obtaining satisfactory quality 
through the application of the same system to 16-mm. film appeared 
so difficult that only a few of the more courageous and optimistic en- 
gineers interested in the problem gave it very serious attention. The 
best of the film records made on 35-mm. film left much to be desired 
in the way of clarity of speech articulation. Low-pitched sounds 
were reproduced very satisfactorily, but high-frequency sounds were 
so deficient that speech sounded badly muffled. No serious difficulty 
was encountered in constructing microphones, amplifiers, recording 
galvanometers, and loud speakers, whose response to high frequen- 

* Received May 24, 1934. 
** RCA Victor Company, Inc., Camden, New Jersey. 

63 



64 



E. W. KELLOGG 



[J. S. M. P. E. 



ties was sufficient to give reasonably satisfactory sound reproduction. 
The photocell, with its connection to an amplifier tube, was in gen- 
eral responsible for some of the loss of high-frequency response, 
owing to the extremely high impedance of the photocell and the fact 
that a comparatively small capacity constituted such a load on the 
circuit as to reduce materially the voltage applied to the grid of the 
coupling tube. The general availability of screen-grid tubes with 
their much lower effective input capacities, plus the development of 
more suitable lamps and better optical systems and the advent of the 
caesium tube (all of which contributed to increasing the total photo- 
cell output, and thereby made it feasible to load the photocell circuit 
with lower resistances), so helped this situation that the proper per- 
formance of the photocell circuit did 
not long remain a serious problem. 
The great difficulty was that, although 
the sound-track on the film was an 
excellent picture of the low-frequency 
waves, the high-frequency waves were 
partly obliterated by fogging and loss 
of exposure, owing to the necessity of 
compressing these waves into a very 
small space. Thus, at 90 feet per 
minute (the standard speed for 35-mm. 
film), a 6000-cycle wave measures only 
0.003 inch from peak to peak. 

Fig. 1 shows an enlargement of a recording of a 1000- and a 6000- 
cycle sine wave. The process of enlargement has eliminated some of 
the fog with which the clear area between the 6000-cycle waves was 
partly filled in the actual recording. When such a record is run 
through a reproducing machine, the failure of the areas which should 
be black to obstruct the light completely, and of the areas which 
should be clear to transmit all the light, results in reduced light modu- 
lation and reduced output from these high-frequency waves. Some 
of this loss is due to the fact that the light-beam, which should ideally 
be an infinitely thin line where it strikes the film, has to be of the order 
of 0.001 inch thick (measured in the direction of film movement) in 
order to get adequate exposure. This is a considerable fraction of a 
wavelength at high frequency. We might compare the process with 
attempting to paint a picture requiring fine detail but using a brush 
that is several times too big. 2 ' 3 Much of the loss at the high frequency 




FIG. 1. (a) 1000-cycle nega- 
tive; (b) 6000-cycle negative. 



Jan., 1935] 16-MM. SOUND MOTION PICTURES 65 

end of the sound spectrum is due to the actual spreading of the light 
within the film emulsion. If a small area of film is exposed and the 
adjacent area protected (e. g., by a knife-edge in contact with the film), 
the resulting film image will reveal a decrease in exposure as the 
edge of the illuminated area is approached, and a very appreciable 
blackening extending into the area which was under the knife-edge. 
The minute crystals of silver halide in the emulsion scatter the light 
like the particles of a fog, causing it to diffuse in all directions. To 
revert to our simile, we have given our painter not only too big a 
brush, but a piece of blotting paper on which to make the picture, 
and paints that run. The harmful effect of this scattering of the 
light is reduced by the employment of high-contrast films and also 
by fine-grained films in which the light is more quickly absorbed and 
the diffusion does not extend so far. 

Other causes of loss of resolution of the high-frequency waves made 
large contributions to our troubles, and added to the losses for which 
the film was responsible, giving a quality of reproduction, which, al- 
though sufficiently good to start the 35-mm. talking pictures on the 
road to commercial success, was far from satisfactory, especially to 
those who had been striving for better quality in radio broadcast and 
similar work. A 16-mm. amateur film would have to run at not more 
than 40 per cent of the speed employed for 35-mm. film, and the waves 
would therefore be correspondingly compressed in length. Thus 
there would be the same difficulty and losses in attempting to record 
a 2000-cycle wave on 16-mm. film as were encountered in recording a 
5000-cycle wave on 35-mm. film. What hope was there, with such 
handicaps, of attaining any worth-while quality on 16-mm. film? 
At least we appeared to be justified in concentrating our main efforts 
for a while longer on improving the 35-mm. performance. One 
factor, at least, was encouraging. The kettle always feels better 
after noting the blackness of the pot. If the high-frequency response 
from film was poor, so likewise was that of many a successful radio 
set. Not only were these sets seriously lacking in high-frequency 
response, as shown by measurement, but they were provided with 
drastic "tone controls" to reduce still further the high-frequency 
output; and we were advised that people were generally using the 
tone controls to their full extent. Perhaps in view of what people 
were accustomed to in the radio field, we might offer to them what 
we had in 16-mm. sound quality, which, though far short of our goal, 
would nevertheless be accepted. 



66 



E. W. KELLOGG 



PROGRESS TOWARD OBTAINING BETTER FILMS 



[J. S. M. P. E. 



In point of time, the development of the 16-mm. sound picture 
projector, in many of its features, preceded the most important im- 
provements in sound recording, but our story will be more connected 
if we continue with the story of the films. 

It is obvious that the loss of resolution of the high-frequency waves 
due to the spreading of the light within the film emulsion will occur 
both in making the negative and again in printing it, and that 
these losses are added. Fig. 2 shows a negative of a 6000-cycle wave 
on 35-mm. film and prints made from it with various exposures. The 
best print is seen to be inferior to the negative. One of the favor- 
able features of the variable-width type of sound record is that a 
printing operation is not required in order to give correct wave shape 




a b c d 

FIG. 2. 6000-cycle negative (a), and three prints (&), (c), (d). 

(a requirement which does apply to variable-density records). We 
therefore have the choice of playing the negative or the print. As- 
suming that we have a satisfactory source of sound or of electric cur- 
rents representing sound, we might make our films by running each 
film through a recording machine instead of making one negative 
and duplicating by means of a printing operation. This would elimi- 
nate the extra loss of resolution which the printing operation entails. 
It might appear at first thought that such a method of producing 
large numbers of films would be excessively costly, but this is not 
necessarily the case. The principal source of sound pictures for a 
16-mm. library would obviously be films previously made on 35-mm. 
stock. The sound recording on these 35-mm. films would be up to 
normal theater reproduction standards, and would at least not be 
impaired by the handicaps to which the 16-mm. sound records are 
subject, owing to their reduced size. Running a 35-mm. film through 



Jan., 1935] 16-MM. SOUND MOTION PICTURES 67 

a synchronous reproducing machine is not a more serious or compli- 
cated operation than running it through a printer. The cost of a 
complete re-recording channel is not so much greater than that of a 
printer as to be a serious item in the expense. It was planned that 
when production of 16-mm. films reached a point to justify the equip- 
ment, we should build re-recording systems in which several 16-mm. 
recording machines are operated in synchronism, all being supplied 
with voice currents from the output of the one 35-mm. reproducing 
machine. Even if it should prove that this method of producing 
duplicate 16-mm. sound films was appreciably more expensive than 
contact printing of the sound-track, many of our engineers felt that 
it would be justified because of the better quality of sound attain- 
able. Those who felt that the very best that was possible would be 
scarcely good enough, were especially inclined to this point of view. 
The amplifiers for the re-recording channels were constructed to 
compensate, at least in part, for the inevitable loss of high frequen- 
cies which the recording and reproduction of sound from the 16-mm. 
film involves. 

For making the pictures on the 16-mm. film, we could start either 
with a 35-mm. negative and run each 16-mm. film through an optical 
reduction printer; or we could start with a 35-mm. master positive, 
make a 16-mm. negative from it by means of the optical reduction 
printer, and then make the 16-mm. duplicate prints by contact print- 
ing. Direct optical reduction of each 16-mm. film gives a distinctly 
superior picture, especially with respect to graininess. Here again 
the better product would cost slightly more, for the reason that the 
optical step printers run more slowly than continuous contact print- 
ers. The operation of printing is itself a small fraction of the total 
cost of turning out the 16-mm. film. 

In the initial work of building up a library of 16-mm. films, the 
system just outlined was followed: namely, each 16-mm. film was 
run through the re-recording machine and then through the optical 
reduction printer. This gave the best 16-mm. film which it was 
possible to make. The sound quality which it was found possible to 
obtain in this way justified full confidence in ultimately developing 
a wide field of application for 16-mm. pictures with sound on the film. 
In fact, the limitation in quality was in considerable measure set by 
the 35-mm. films which we were able to obtain at the time. 

Intensive study was given to the characteristics of films and de- 
velopers and of the effects of various exposures and developments. 



68 E. W. KELLOGG [J. S. M. p. E. 

The outcome of these studies was that no films* were immediately 
available which were better, all things considered, than the Eastman 
positive with which we had started, nor was any developer found 
which was better than the Eastman D-16. Equivalent film emul- 
sions were obtained from other manufacturers, and faster films with 
substantially the same resolving power have since been developed. 
The standard cine-positive films are, in fact, very well suited to sound 
recording. They are the finest grained of all of the commercial films, 
and about equivalent in this respect to special process films, which 
are rated as extremely fine grained. They are also inherently high- 
contrast films, a factor which is favorable in variable-width recording, 
since this high contrast helps enable us to obtain deep blacks and 
clear whites with relatively sharp lines of demarcation between them. 
A very complete study of printing and development was made. It 
appeared in general that a negative from which a print is to be made 
should have a density in the black portions (density being defined 
as log 10 of the ratio of incident to transmitted light) of the order of 
1.8, preferably obtained with a development giving a "gamma" or 
contrast factor of 2 to 2.2, while the best prints obtained from such 
negatives had densities of 1.3 to 1.5, with a gamma of about 2.0. 
Negatives with lower contrast (although they would in some cases 
give prints having as great a high-frequency output) gave less lati- 
tude in printing, and required lighter prints for maximum print out- 
put, which is disadvantageous from the ground noise standpoint. 
Thinner or lighter negatives also called for light printing. Shorter 
print development was found unfavorable in permitting less latitude 
in print density, and requiring lighter prints for equal high-frequency 
output. The best density for a recording to be played directly was 
found to be of the order of 1.0 to 1.3 (5 to 10 per cent of the incident 
light transmitted), obtained with a development to a gamma of 2.0 
to 2.2. 4 In the course of the experiments with various film emul- 
sions, several were tried which had been panchromatized in order to 
make it possible to obtain the necessary exposure more readily with 
incandescent tungsten light sources, which are somewhat deficient 
in blue light. In all these tests, it appeared that the sensitiveness 
of the panchromatized films to red and yellow light was obtained at 
some cost in resolution. This is, no doubt, in part due to the fact 

* Recently, certain still finer-grained films have shown their superiority, but 
they have not as yet been put to commercial use. 



Jan., 1935] 16-MM. SOUND MOTION PICTURES 69 

that the blue light is more quickly absorbed in the emulsion than 
light of greater wavelength, and therefore does not spread as far. 
A film whose exposure depends almost entirely on blue and violet 
light gives the best resolution. It is possible to go further in this 
direction by the employment of yellow dyed film (duplicating stock). 
Although excellent resolution was obtained, the yellow dyed film is 
of inherently low contrast and this factor is unfavorable, so that the 
net result was not better than that obtained with standard positive 
stock. 

Improvements in optical systems contributed in an important way 
to making better 16-mm. sound-films. The fine line of light cast 
upon the film for either recording or reproduction is an optical image 
of a slit between metal plates. Although the standard 16-mm. 
microscope objective was not designed for this purpose, it was found 
from the start to be admirably suited ; and tests with a large number 
of other lenses, some of them specially designed and corrected for 
minimum aberration for blue light (instead of for the colors for which 
the eye is most sensitive), revealed no lens appreciably superior to 
the standard 16-mm. 

The most important improvements were in the recording optical 
systems, especially in the galvanometer. The first Photophone re- 
cording systems employed practically standard oscillograph vibrators 
with mirrors 0.017 by 0.060 inch. In any oscillograph and in a re- 
cording optical system employing an oscillograph vibrator, light is 
scattered at the surfaces of the lenses and by the edges of the mirror. 
Specially selected mirrors were employed, and the edges blackened in 
our efforts to decrease the amount of stray light reaching the film. 
The galvanometer suspension strips were also blackened, for the same 
reason; but there was an inherent limitation in the extremely small 
size of the mirror, and it did not appear feasible to increase the size 
materially in a vibrator of the standard type. Work on a galvanome- 
ter of the moving iron armature type had been undertaken by C. R. 
Hanna, of the Westinghouse Electric & Mfg. Company, and carried 
on by G. L. Dimmick, of the General Electric Company and later of 
the RCA Victor Company. 5 A galvanometer was developed having 
a sensitivity about equal to that of the standard oscillograph vibrator 
(same angle of deflection for equal watts supplied), and carrying a 
mirror 0.100 by 0.125 inch, or over ten times the size of that used in 
the former type of galvanometer. Better selection and placing of 
some of the lenses and the development of a better lamp contributed 



70 E. W. KELLOGG [J. S. M. p. E. 

to the working out of an optical system having much less stray light 
than had been previously possible. The reduction in stray light helps 
to avoid fog and loss of contrast in the film record, while the increase in 
the total amount of available light, due again principally to the large 
mirror, makes it possible to obtain adequate exposure with a narrower 
slit image at the film. Whereas the first recording systems produced 
an image 0.001 by 0.060 inch, it is possible with the new optical sys- 
tems to get plenty of exposure with an image only 0.00025 wide. 
It has already been pointed out that the width of this slit image or 
line of light is one of the important factors causing loss of resolution. 
Although the building up of commercial 16-mm. libraries, by the 
process of recording the sound directly on the edge of the film, was 
undoubtedly feasible, commercial film laboratories foresaw difficul- 
ties in introducing equipment of a distinctly different type and call- 
ing for a personnel of somewhat different training from that required 
to operate their standard printing equipment. The fact that it was 
now possible to make better sound negatives rendered it possible to 
revert to the system of contact printing and still obtain prints which 
were considered of adequate quality. Improvements were also made 
in contact printers. 6 The process of direct recording was therefore 
not widely applied. Another method of making the 16-mm. prints, 
however, has shown even better results than the direct recording; 
and since it is purely a printing process, it is not open to the objec- 
tions made to the direct recording, namely, to optical reduction of 
the sound-track from 35-mm. negative. 7 ' 8 Early tests on optical 
printing had appeared to indicate that it was slightly inferior to the 
best contact printing. There are two factors, one favorable and one 
unfavorable. In a contact print all the light which gets through the 
film is effective in exposing the print. In an optical printing system, 
on the other hand, the lens subtends only a small solid angle from the 
negative and collects only the light which emerges in nearly the same 
direction as it strikes the film. In other words, the lens receives 
little except specularly transmitted light, and the scattered light is 
discarded. Since the fraction of the light which is scattered is small 
in the thin portions, and much greater in the denser regions, the re- 
sult is that an optical printing system increases the contrast as com- 
pared with contact printing, and increased contrast is desirable in 
variable-width sound recording. Imperfections in the optical sys- 
tem, on the other hand, tend to cause loss of resolution, and the suc- 
cess of an optical printing system as compared with contact printing 



Jan., 1935] 



16-MM. SOUND MOTION PICTURES 



71 



would depend on minimizing these optical imperfections. The in- 
equality in the reduction in the lengthwise and crosswise directions 
calls for a special optical system. The width of the sound-track on 
a 35-mm. film is 0.070 inch while that on the 16-mm. film is 0.060 inch. 
Thus a reduction in the ratio of 6 to 7 is called for in the transverse 
direction. In the other plane, the reduction should be in the same 
ratio as the relative speeds of the films, in order that the image of the 
negative may move exactly with the 16-mm. film. This calls for a 
2.5 to 1 reduction. Several optical reduction systems were worked 
out, employing cylindrical lenses or combinations of cylindrical and 
spherical lenses. In these, it was important to employ the cylindri- 



































































































































^ 


^ ^ 





















/i 


C 


A 


TtVf A 


MD 


Offf 


?/? 


EC 


0, 


n 


>VG 


/* 


r /t 


:/ 


t 


S 


\. 
^\ 














^ 


so 


t* 


t 


* 


CONS 


r AtfT 


GA 


LV 


At 





M 


-r t 


y 


At 


1, 


6/ruo 


V 


\ 


\ 

\> 












4 






































\ 


^ 










Q 


6O 


^c 


p 


l~j 


CAL I 


lDi 


icr 


/O, 


Y t 


/ 


/, 


'/ 


/?< 


7/1 




f 5 MM 


F/L 


^ 




\ 


^ ^ 








n. 




/ 


A 


V 


HG C 


OHS 


TA, 


vr 





(/ 


r / 


L'T 


















\1 








^ 










































s 


\ 






s 




3~J 


>/? 


It 


<ro/v(. 


~*OMfi 


fff 


C/A 


I ( 


\ 


* J 


i v. 


fi 


/A 


Ti 


ifevSE 


I AS 


Vfc 


AT 


fV 


t) 




V 




<!: 


?0 














































2 

























































IOO CYCLC3 PfftSfCOflO 'OOO /OOOO 

OUTPUT of /6n F/LMS 



FIG. 3. Characteristics of fiJms made in different ways. 

cal lenses (which are available at present only in very simple form) in 
such a manner that then* imperfections resulted in least impairment 
of the image, while the image in the critical direction (line width) was 
formed primarily by the highly corrected spherical lens. Variations 
in the film length due to shrinkage would result in slight departures 
from the 2.5 to 1 speed ratio, but if the image has to follow for an ex- 
tremely short distance, the slight creepage of film with respect to the 
image cast upon it becomes negligibly small. The optical systems 
employed gave sufficiently intense light so that exposures could be 
obtained while the 16-mm. film travels 0.005 inch or less. The factor 
which gives optical sound printing of 16-mm. films from 35-mm. 
negatives a big advantage over contact printing is that the printing 



72 E. W. KELLOGG [J. S. M. P. E. 

can be done with an almost perfect negative. On the 16-mm. film 
there does not at present appear to be much hope of usefully re- 
cording frequencies higher than 4000 or 5000, and in records of 
these frequencies, the fogging is such as to cause a loss of output of 
the order of 50 per cent. On the other hand, a 35-mm. film does not 
encounter the same difficulties until a frequency of the order of 10,000 
is reached, while at 4000 cycles a well recorded 35-mm. negative shows 
practically no fogging or filling in. The difference is qualitatively 
illustrated in Fig. 1. It was therefore possible to project upon the 
16-mm. raw stock an image of a 4000-cycle wave which had almost the 
quality of the 35-mm., 4000-cycle negative; whereas, with a contact 
print the image could be no better than the 16-mm., 4000-cycle nega- 
tive. Fig. 3 shows the output obtained from 16-mm. films produced 
by optical and contact printing and by direct recording. It is seen 
that the optical printing gives even better results than the direct re- 
cording. It is, of course, not possible to introduce any compensation 
for loss of high frequencies in the optical reduction printing, but it 
has been found unnecessary if we start with a really good 35-mm. 
film; and it is also better to reduce an imperfection or loss than to 
compensate for it. Should it appear desirable in special cases to 
introduce compensation, this is still possible by making a new 35-mm. 
negative by re-recording from the original. 

DEVELOPMENT OF A PROJECTOR 

The popularity of "talkies" in the theaters soon led to endeavors 
to provide sound for 16-mm. pictures. As was the case with the 
35-mm. film, the first equipments offered to the public employed disk 
records. The advantages of having the sound on the film, however, 
were so great that it needed only to be established that adequate 
sound quality was attainable to cause the 16-mm. development to 
follow the same path and abandon the disk. 

As soon as the more pressing problems in the development of 
35-mm. sound systems had been worked out, a number of engineers 
undertook the development of 16-mm. sound pictures with the sound 
recorded on the edge of the film. Reproducing optical systems and 
photocell arrangements already developed for 35-mm. films were, 
of course, applicable to 16-mm. projectors, and there was not a great 
deal of chance for improvement in the optical systems so far as per- 
formance was concerned. Developments centered rather around 
various practical expedients for simplification or reduction in size. 



Jan., 1935] 16-MM. SOUND MOTION PICTURES 73 

Only a few of the many problems encountered and expedients em- 
ployed will be mentioned. 

Before a projector could be built, it was necessary to decide where 
to put the sound-track. Simply scaling everything down in the 2.5 
to 1 ratio is the most obvious expedient; but it would be very unde- 
sirable to narrow the track in the same proportion as the picture 
frame size because the amount of light which we can send through 
the 35-mm. track is none too much, and narrowing the track would 
mean a proportional reduction. Narrowing the track also aggra- 
vates the ground noise problem. Furthermore, tolerances in side- 
wise play of the film 9 can not practically be reduced very far below 
those necessary for the 35-mm. film, so that either the sound-track 
or the picture, or both, would have to be narrowed down more than 
in proportion to the available width between sprocket holes, and the 
change in shape and size of the 35-mm. picture frame which had been 
made to accommodate the sound-track was conceded to have been 
unfortunate. The alternatives were to adopt a wider film, or else to 
omit one row of sprocket holes. The adoption of a wider film would 
have meant that our machine could not project present standard 
silent pictures. The omission of one row of sprocket holes was there- 
fore the preferred solution, provided it did not risk reliability or film 
life. The success of a number of commercial cameras and projectors 
whose intermittent claw movements engaged the perforations only 
on one side was sufficient answer to this question; and with further 
encouragement by experienced engineers of the film companies, our 
engineers adopted this plan. 

An interesting solution was found for the problem of lighting the 
exciter lamp without introducing hum. Engineers in sound picture 
work are quite familiar with the fact that lighting the lamp of a photo- 
electric reproducing system with alternating current results not only 
in hum but an objectionable modulation of the audio frequency out- 
put, although the fluctuations in light are entirely too small and rapid 
to do any harm for visual purposes. The use of steady direct cur- 
rent, of course, avoids this trouble. Batteries were not to be con- 
sidered. Lamps suitable for reproducing systems call for low voltage 
and relatively high current. Rectifying and filtering the supply 
would call for extra equipment, which would be rendered all the more 
bulky and heavy because of the difficulties of filtering a low- voltage 
high-current supply. On the other hand, the rectifying and filtering 
system already required for the amplifier could supply the additional 



74 E. W. KELLOGG [J. S. M. p. E. 

power needed for the exciter lamp, but at high voltage and low cur- 
rent. This high-voltage direct-current supply furnished the plate 
circuit power for a 10,000-cycle oscillator. An extremely simple 
and light 10,000-cycle transformer steps down the voltage to a value 
suitable for the exciter lamp. 

COMPACT OPTICAL SYSTEM 

The optical system employed for 35-mm. projectors is objection- 
ably large and calls for a 75-watt lamp, which presents a heat dissi- 
pation problem as well as calling for a large amount of power. These 
factors are of no serious consequence in a theater installation, and the 
35-mm. optical system, although very inefficient in utilizing light, 
represents about the best we know how with regard to light intensity 
and perfection of image. A system in which no slit is used, but in which 
the lamp filament itself is imaged upon the film, is far more economi- 
cal. One of the arrangements which had some time ago received 
consideration was to employ a straight filament and image it by a 
standard spherical lens upon the film. It is practically impossible, 
however, to meet the requirement that the filament should not devi- 
ate from a straight-line perpendicular to the direction of film travel 
by more than a small fraction of its own diameter. A cylindrical 
lens, however, will produce a straight line of light from any concen- 
trated source. In the present case, the lamp has a helical filament, 
approximately 0.008 inch outside diameter, imaged upon the film by 
a cylindrical lens, with a reduction ratio of Vio- The length of the 
helix parallel to the axis of the cylindrical lens influences only the 
brightness of the line of light. The hot part of the helix is about 
Vie inch long. If the helix is not parallel to the cylindrical lens axis, 
the image is widened. It has been found feasible to obtain lamps in 
which the variations in filament angle are well within the limits re- 
quired to prevent serious widening of the image. Following a sug- 
gestion made by Prof. A. C. Hardy, in connection with a different 
optical problem, advantage has been taken of the fact that the 
cylindrical lens may be of shorter focus and placed closer to the film 
than would be feasible with a spherical lens system, and the impair- 
ment of image quality as a result of lens aberration is corre- 
spondingly reduced. It is difficult to get cylindrical lenses in any 
but the simplest form. The lens we are using for these projectors 
consists simply of a short cylindrical glass rod, one side of which is 
ground flat. In spite of the crudity of a lens of this type, deviations 



Jan., 1935] 



16-MM. SOUND MOTION PICTURES 



75 



of the light rays from the ideal directions do not spread the image 
appreciably, because the misdirected light rays have such a short 
distance to go. The employment of a very short focus cylinder 
means that the lamp is also brought moderately close to the film and 
the entire optical system is reduced in dimensions. Fig. 4 shows 
approximately the relative size of the 16-mm. and 35-mm. optical 
systems. The lamp operates on 3 /4 ampere at 4 volts. It is, of 
course, not feasible to operate a lamp of this rating at as high a 
temperature as the 10-ampere lamp used in the larger system, and 
the light intensity is correspondingly reduced, although this loss 



FILAMENT 



PHOTOCELL 




VERTICAL 
CYLINDERS 

3 W. LAMP 

FIG. 4. Relative sizes of the 16-mm. and 35-mm. optical systems. 

has not proved serious. An amplifier system has been worked out 
which is built into the projector unit. The amplifier is so com- 
pact that it hardly necessitates any increase in the size of the case 
above what would be required for the projector itself. 

Self -threading has been advocated by many for amateur projectors. 
Several successful self -threading laboratory models were worked out, 
employing a number of ingenious devices, but self -threading inevita- 
bly means more mechanism, less accessibility, and usually special 
reels, and the argument for it is based upon the supposition that 
manual threading is difficult or tedious. Making the manual thread- 
ing simple and easy is, we believe, a better solution. Self-threading 
machines have been available for silent pictures for some time but 
have not been in great demand, and the situation is not made ma- 
terially different by the addition of sound, for anyone who can thread 



76 E. W. KELLOGG [J. S. M. P. E. 

film into a picture gate with proper loops can certainly perform the 
much simpler operation of threading a sound element. 

The form in which a 16-mm. sound picture system should be offered 
to the public has been the subject of much difference of opinion. The 
idea that a complete equipment should be in one cabinet which should 
be a handsome piece of furniture makes a strong appeal, and a number 
of manufacturers have built equipment along these lines principally 
employing disks for sound reproduction. While there is unquestion- 
ably a field for units of this kind, they would necessarily be consid- 
erably more expensive and have several serious limitations. A 
single-unit device takes up considerable room and may present more 
serious problems to the housekeeper than a three-unit outfit (pro- 
jector, loud speaker, and screen) which can be readily set up when 




FIG. 5. Complete 16-mm. reproducing equipment. 

wanted and be taken away between times. The single-cabinet 
equipment would also call for a short-throw projection system, and 
the picture size is limited. One of the most serious objections to the 
one-cabinet system is that the inevitable projector noise comes to 
the listener from the same direction as the sound from the loud 
speaker. We have all observed how much more readily we may con- 
centrate upon a desired sound and ignore a disturbing noise if the 
latter comes from a different direction. The equipment described 
here is thus designed for giving the best possible show, and the fact 
that it must be set up when wanted and taken down between times is 
compensated by the simplicity of setting it up. Fig. 5 shows the 
projector and loud speaker cases.* A very convenient and efficient 

* A new model is now available having all the features discussed above, 
but with increased screen illumination and sound output as well as a num- 
ber of other improvements. 



Jan., 1935] 16-MM. SOUND MOTION PICTURES 77 

screen is provided which can be set upon a table and unrolled. The 
loud speaker unit should be placed as close to the picture as possible, 
usually directly below but somewhat raised from the floor. A 25-ft., 
4-conductor cable carries the field and voice currents from the pro- 
jector-amplifier unit to the loud speaker. The size of the speaker 
box is dictated by acoustical considerations. The spare space in the 
speaker box is made available for carrying films, cable, and spare 
tubes. 

REAL HOME MOVIES 

Important commercial development of 16-mm. sound pictures 
would be assured even though all the films had to be taken originally 
on 35-mm. film and then reduced to 16-mm. The cost and compli- 
cation of this method of making films would be justified in view of 
the quality of films that can be made by this process for film libraries 
(most of the subjects for which are already available on 35-mm. film) 
for entertainment and education and for commercial demonstration 
and advertising. The real fun in home movies, however, would be 
for the amateur to make his own pictures, with his own family and 
friends as the actors, or for the traveller to record his thoughts and 
impressions while he takes pictures of various scenes. 

Considering the formidable array of equipment which motion pic- 
ture producers have had to purchase or lease to do their work the 
staffs of picture and sound experts, the gigantic sound stages with 
elaborate sound-proofing and tons of sound absorbing material 
what hope could there be of supplying the amateur with simple and 
cheap equipment of extreme portability, with which he could, with- 
out any other special facilities, take satisfactory talking pictures of 
inexperienced actors? 

No story of the development of amateur sound pictures by the 
RCA Victor Company would be complete without recognition of the 
important part played by C. N. Batsel. 10 While his associates were 
struggling with making 16-mm. films by reduction from 35-mm. 
records, Mr. Batsel, feeling that the 16-mm. system would never be 
complete until people could take their own home movies, had visions 
of a complete equipment so simple, compact, and portable that its 
possession and operation would be within the reach of ordinary non- 
professionals. Without the encouragement, or much optimism on 
the part of others, Mr. Batsel began working with an adaptation of 
an existing Bell & Howell camera. Fitting a sound system into a 



78 E. W. KELLOGG [J. s. M. p. E. 

camera not designed for the purpose involves either compromises or 
greater difficulties than making a completely new design ; but in spite 
of this Mr. Batsel, with his roughly modified Bell & Howell camera, 
made sound recordings of sufficient quality to make us all recognize 
that the objective was by no means unattainable. On the strength 
of these results, a complete camera and recording system was con- 
structed, which brought the project a step nearer to realization. The 
recording camera was in the form of a brass box about 4 by 8 by 10 
inches, weighing about 15 pounds. Amplifier and battery were 
carried in a case about 6 by 12 by 18 inches, and a camera tripod, a 
collapsible musician's stand for microphone support, and about 25 
feet of three-conductor cable completed the equipment. One man 
could carry it all, and set it up for operation in less than two minutes. 
The camera and recording element were driven by a standard phono- 
graph spring, and a phonograph governor was employed to control 
speed. The recording optical system was the same in principle as 
the one Mr. Batsel had used. A 3- watt lamp of the kind used in the 
projectors has its filament imaged by a small condenser lens on the 
mirror of a magnetic galvanometer; and this, in turn, is imaged on 
the film by a Vs-inch focus cylindrical lens to form a line of light 
about 0.0008 inch wide. Rotation of the mirror shifts the end of 
this line of light, thus exposing more or less of the width of the sound- 
track. The system is like that of a standard oscillograph, but on a 
very small scale. 

The brass box camera equipment served to try out the field of 
amateur movies to ascertain how serious were the acoustical prob- 
lems, how consistently the equipment would work, and whether ex- 
traordinary skill would be required. One of the factors that turned 
out most fortunately was that for outdoor pick-up the only difficulty 
was that of finding adequate freedom from disturbing noises. The 
huge specially constructed stages of Hollywood are necessary for 
making indoor pictures, at any time of day or night, independently 
of what is going on in the neighborhood. But in a quiet suburban 
or country district, a still more perfect sound stage (acoustically) is 
obtainable without building anything at all. Whereas in an ordinary 
room the microphone can be hardly more than from one to two feet 
away from the speaker without ruining the record, we could get dis- 
tinct speech in a quiet place out-of-doors at distances up to 15 feet 
or more. A microphone lacks the power of concentrating upon 
sounds coming from certain directions which a human being with 



Jan., 1935] 16-MM. SOUND MOTION PICTURES 79 

normal hearing possesses, and must be placed much closer to the 
sound source than one would expect in view of ordinary listening ex- 
perience. Satisfactory recordings were made out-of-doors so long 
as the sound reaching the microphone was loud enough to give good 
modulation on the film. If the persons taking part could remember 
not to let their voices fall to too low levels, the microphone could be 
placed far enough away to be out of the picture. Special interest 
is attached to making sound pictures of children. Children's voices 
are, in general, harder to record than adult's voices, and it is more 
difficult for them to keep between reasonable limits of loudness and 
faintness. A number of excellent pictures of children were made, 
however, in which the parents, at least, have taken extreme delight. 
The experience gained with the brass box camera model was suffi- 
cient to demonstrate beyond question that the amateur sound pic- 
ture was practicable. Up to this point the films had been developed 
as negatives and contact printed, but for general applications it was 
felt that the reversal process should be followed. The recording of 
sound as well as pictures upon a reversal film brought new require- 
ments, but after extended tests the Eastman Kodak Company, whose 
cooperation in this development has been very complete, found it 
possible to furnish a reversal stock and process it so as to give the 
same excellent pictures that are possible in silent motion pictures, and 
a sound-track superior to that obtained with a negative and print. 
Fortunately for the amateur, the singly perforated reversal sound 
stock with its processing will cost him no more than the present silent 
pictures. 

AUTOPHONE CAMERA 

Although there was no longer any question as to the practicability 
of amateur sound pictures, there were differences of opinion as to 
what would be the most important field for such equipment. It was 
held by some that the greatest demand would be on the part of people 
taking trips or travelling in foreign lands, not necessarily to record the 
sounds made by the subjects of the picture but to record their own 
comments, descriptions, and explanations. Mr. G. L. Dimmick had at 
this stage of development demonstrated an extremely simple device 
with which he had been experimenting, in which the mirror was caused 
to vibrate by connecting it directly to an acoustically actuated 
diaphragm. In this Mr. Dimmick revived the principle of C. A. 
Hoxie's "pallophotophone," which R. P. May had proposed to use 
in our portable sound recording equipment. Such a device might 



80 E. W. KELLOGG [J. S. M. P. E. 

be substituted for the magnetic galvanometer which we had been 
using, and would of course require no microphone or amplifier. This 
appeared to offer the possibility of a still simpler, lighter equipment 
for use by travellers and for similar service. It was necessary, of 
course, in order to impart adequate movement to the mirror, to place 
the mouth quite close to the diaphragm as in talking into an ordinary 
telephone transmitter, and this precludes recording the speech of the 
person in the picture; but a recording camera might be constructed 
in which the mouthpiece was in the back of the camera and the op- 
erator could himself talk while looking through the finder and run- 
ning the camera. 

The development of a recording camera for this service brought in 
a number of new problems. In the first place, the person using the 
camera would not wish to bother with the tripod or to stoop over 
to the level of the tripod, as would be practically necessary if he em- 
ployed a tripod. The camera must therefore be light, convenient to 
support in one's hands and hold before one's face for a reasonable 
time without causing weariness. It must also be small enough not to 
occupy undue space in one's baggage. Complete redesign of the 
camera mechanism was necessary to meet these new requirements. 
The utmost simplification of mechanism and lightening of parts was 
necessary. A number of the detail problems will be mentioned. 
Obviously, if a sufficiently light and compact camera could be worked 
out for this autograph service, the same general design might be em- 
ployed, with the substitution of a magnetic galvanometer and the 
addition of a separate microphone and amplifier, for general sound 
picture work. 

Messrs. G. L. Dimmick and L. T. Sachtleben undertook the fur- 
ther development of the acoustical recording device or "autophone 
element." One of the problems consisted in causing very small 
movements of the actuating device to result in relatively large rota- 
tions of the mirror. The same problem exists in the magnetic gal- 
vanometer, which for this purpose employs a knife-edge on the arma- 
ture, registering with a groove in a bar which is pivoted by means of 
strips of phosphor bronze under tension, as shown in Fig. 6. The 
closer together the stationary and moving pivots or hinge points, the 
greater will be the mirror rotation for a given armature movement. 
The center of the effective pivot produced by the bronze strips is 
only 0.020 inch from the knife-edge, and at this distance a movement 
of Veooo of an inch results in a rotation of 0.5 degree, which, in the 



Jan., 1935] 



16-MM. SOUND MOTION PICTURES 



81 



optical system employed, is sufficient to give full modulation of the 
light on the film. In the case of the autophone, a double knife-edge 
construction was employed, the bar which carries the mirror being of 
a diamond-shaped cross-section, with the two edges 0.020 inch apart. 
The mirror is 0.100 X 0.125 inch. In order to secure adequate sen- 
sitivity, it was of course necessary to make the diaphragm very 
light; but at the same time the central part must be stiff enough to 
transmit the forces applied by air pressures from all parts of the dia- 
phragm to the pin which rocks the mirror, and the diaphragm must 
be reasonably rugged. Aluminum alloy diaphragms of various thick- 





(A) 
RELAT 

(A) 3\ 
Oi 
o.O 
(B)Nt 




POLC Piece 

ARMATURE\I 
' '^'-^ /* 


i 
i 

N 


--I 

MAGNET i 
*u \ 




X 


h 


CB) ^ 
fve MIRROR *'**lJ|i\ 


r ANDARD U\ 
CILLOGRAPH ^N 

/ 7*0.060" BnoNze STRIP ^ 
~w GALVANOMETER 


S 


X 


5 




1 

MAGNET 
POLE 



FIG. 6. Features of magnetic galvanometer. 



nesses and forms were tried, a diaphragm finally being chosen of 
0.0016-inch aluminum alloy, the central portion being stiffened by 
concentric corrugations, while the outer edge is made as flexible as 
possible. Such a system necessarily shows a fundamental resonance, 
the frequency of which is determined by the diaphragm stiffness and 
the mass of the diaphragm plus its loading by the mirror. Damping 
of the resonance is obtained by inclosing the air-space behind the 
diaphragm, except for suitably sized holes covered with a fine porous 
paper. It was found possible to build up the sensitivity of the de- 
vice in certain frequency ranges by utilizing broad acoustical reso- 
nances. Thus the cavity in front of the microphone can constitute a 
resonance chamber, with resultant building up of pressure in the 
neighborhood of the resonance frequency, while the opening of a 



82 E. W. KELLOGG [J. S. M. p. E. 

small air-passage between the front and back cavities gave a reso- 
nance in which the maximum pressure variations occur in the back 
cavity at a much lower frequency. 

When one speaks as close to the mouthpiece as is necessary for a 
device of this kind, puffs of breath produce pressures which are large 
compared with those due to actual sound. It was necessary to place 
screening in the mouthpiece and to vent the space behind it in order 
to prevent the light from being blown off the sound-track, especially 
following explosive consonants like P and B. The screens also serve 
to prevent the moisture in the breath from collecting on the dia- 
phragm. In view of the extremely small movements of the dia- 
phragm required to throw the light spot all the way across the sound- 
track, special precautions were necessary to prevent shifting of the 
zero position (which should be such that half the track width is il- 
luminated). Inequalities in thermal expansion, for example, may 
readily throw the light spot off the track, or at least seriously shift 
its zero position. To prevent this, it was found necessary to con- 
struct not only the diaphragm, but its housing and other parts of the 
structure, of the same aluminum alloy. This measure alone would 
not have prevented sudden warming of the diaphragm from the 
breath with consequent shifting of the light spot, but the mouthpiece 
screen was evidently effective in preventing appreciable warming of 
the diaphragm by the breath at a rate faster than could be cared for 
by the conduction of the heat through the diaphragm to its support. 
A small vent between the screen and the diaphragm proved effectual 
in preventing trouble from the puffs of breath. 

Both sides of the diaphragm must be shielded as well as possible 
from noise produced by the camera mechanism. To this end the 
entire autograph element is housed in its own casting, which com- 
pletely encloses the space behind the diaphragm. Careful design 
and close tolerances help to minimize the noise of the intermittent 
movement. 

OPTICAL SYSTEM ADJUSTMENTS AND MONITORING 

Although it is not expected that adjustments will often be neces- 
sary, it is practically necessary to make the galvanometer adjustable, 
so that the light spot may be properly centered upon the sound- 
track. In the case of the magnetic galvanometer, this adjustment 
is accomplished by rotating the entire galvanometer. With the 
autophone unit this is not feasible, and the adjustment is afforded by 



Jan., 1935] 16-MM. SOUND MOTION PICTURES 83 

making the groove which registers with the stationary knife-edge 
capable of being moved a very small amount. 

Pushing a small shield aside closes the lamp switch and exposes a 
peep-hole in the side of the camera, through which the light spot can 
be seen where the light strikes the cylindrical lens aperture. Marks 
on the aperture plate show when the light beam is properly centered 
and how far it must move for full modulation. The person operat- 
ing the autophone camera can readily have some one check his voice 
level by observing the amplitude of movement of this light spot while 
he speaks into the mouthpiece. If the microphone is being used, the 
operator can check the recording level (which is controllable at the 
amplifier) by watching the movements of the light spot. An addi- 
tional means of monitoring is provided in the form of a tiny neon 
lamp on the amplifier, which flashes whenever the voltage applied 
to the galvanometer exceeds a certain value. The purchaser of a 
camera is instructed to adjust the loudness level, by properly coach- 
ing his subjects and by adjusting the microphone distance or the 
amplifier volume control, so that the neon lamp flashes occasionally. 
A telephone receiver is also furnished with the microphone equipment, 
so that the operator can judge the quality of the sound being recorded. 
In this connection, it may be said that (although the device is capable 
of working satisfactorily over a reasonable range of loudness), as in 
all sound recording, the results are best when the record is made at 
the maximum loudness level which can be accommodated without 
overshooting. This does not mean shouting, but it means especially 
the avoidance of the tendency which many speakers have, of uttering 
some syllables quite loudly and others faintly. If the person making 
the record can remember to hold a fairly sustained and uniform loud- 
ness and to speak distinctly and moderately slowly, he may feel con- 
fident of an excellent record. 

One of the most important adjustments in any photographic sound 
recording system is the focusing of the line of light on the film in 
order that it may be as fine and sharp as possible. With the optical 
system employed, the tolerance in the distance which a given lens 
may be from the film without appreciable impairment of the recording 
is about plus or minus 0.002 inch. Once a camera has been assembled 
and properly adjusted, however, there should be no necessity for re- 
adjustment. The arrangement for focus adjustment has therefore 
been designed with a view to production, and checking, if necessary, 
by a properly equipped service man, rather than for ready accessi- 



84 E. W. KELLOGG [J. S. M. p. E. 

bility on the part of the owner. The optical system is focused in a 
jig in the factory, and an adjustable stop pushed into contact with a 
boss on the jig, and clamped. When the optical system is placed in 
the camera, the adjustable stop is placed against the sound sprocket 
bearing bushing, which has an accurate position relative to the sur- 
face of the sprocket. The screws which hold the optical system in 
place are then tightened, with the distance between the objective 
lens and the surface of the sound sprocket accurately determined by 




FIG. 7. Autophone camera, showing mouthpiece 
and finder. 

the setting of the adjustable stop. With this arrangement, optical 
systems are interchangeable in cameras, and changing from auto- 
phone unit to magnetic galvanometer or vice versa does not upset the 
focus adjustment. 

Fig. 7 is a view of the autophone camera. The finder and mouth- 
piece are so located that when the operator is looking into the finder 
his mouth will be opposite the mouthpiece. Just to the right of the 
mouthpiece and optical system is a compartment containing the 
three No. 2 dry-cells required for lighting the exposure lamp. This 



Jan., 1935] 16-MM. SOUND MOTION PICTURES 85 

autophone camera is without doubt the smallest complete picture 
and sound recording system in existence. 

GENERAL FEATURES OF THE RECORDING CAMERA 

Fig. 8 shows the sprocket and gate arrangement and the path of 
the film through the camera. The picture gate design is such as to 
make threading easy, and the shoe can be readily removed to inspect 
or clean the surfaces. There are no movable rollers or guides, but 
the film is slipped into place between stationary rollers and the 




FIG. 8. Sixteen-mm. sound camera, autophone type with 
covers removed. 

sprocket. With the rollers properly placed, we found that this ar- 
rangement resulted in greater convenience for threading than a mov- 
able roller system. 

It is obviously important that the film remain snugly against the 
sprocket at the exposure point. If the film were under considerable 
tension, this would not constitute a problem; but where there is 
little tension on the film, which is a desirable condition for sound 
sprockets, faulty design often results in a tendency for the film to 
loop away from the sprocket. A careful study was made of this sub- 
ject. It was found desirable to restrict the angle of contact to about 



86 E. W. KELLOGG [J. S. M. p. E. 

90 degrees, and to bend the film back around the guide rollers so that 
the natural stiffness of the film would tend to hold it against the 
sprocket. It is this utilization of the natural stiffness of the film 
which makes it possible to do away with movable rollers. With the 
arrangement chosen and with a properly dimensioned sprocket, the 
tooth action is very smooth. 

The camera can be loaded in daylight, using the same arrangements 
that have been employed for the purpose with silent cameras. The 
take-up reel is driven by gears, with a slipping clutch arrangement 
which gives more satisfactory action than a spring belt, which has an 
annoying habit of unwinding some of the film after one has turned 
the reel to take out the slack. 

There are numerous types of friction governors which will hold 
speed sufficiently closely for silent picture work, but no other type 
was found comparable with the standard phonograph governor 
where the close speed regulation required for sound recording is 
needed. In order that constant speed of the governor may mean 
constant speed of the film, it is necessary that the sound sprocket be 
accurate and that the gearing between the two be as simple as pos- 
sible, as well as accurate. A large diameter gear on the sound- 
sprocket shaft meshes with a pinion on the governor shaft, and the 
mounting of the pinion is designed to minimize possibilities of eccen- 
tricity. The low-speed gearing between the main-spring and the 
sound sprocket has not been found to be a source of disturbance, but 
applies a very steady driving force to the sound-sprocket shaft. The 
gears on the sound-sprocket shaft are part of the train which drives 
the highest-speed element, namely, the intermittent movement, and 
it is necessary to insure that the load of the intermittent shall not re- 
act appreciably on the speed of the sound sprocket. The best solu- 
tion of this problem was found to consist in providing an extra gear 
on the sound-sprocket shaft which meshes with a pinion on the in- 
termittent shaft, this extra gear being loosely mounted on the sound- 
sprocket shaft and driven through a spring. The spring incidently 
serves to absorb the shock from the governor when the camera is 
suddenly stopped. 

Various types of footage counters were tried in preliminary models. 
The type finally chosen was one which, in effect, measures the radius 
of the roll of film on the supply reel. Although other types of footage 
counter may give more accurate readings, this has the merits of reli- 
ability and simplicity. Most other types have to be reset when a 



Jan., 1935] 



16-MM. SOUND MOTION PICTURES 



87 



new film is put in the camera. A good feature of this type of footage 
counter is that the accuracy of the indication becomes greater to- 
ward the end of the reel when the operator must watch his available 
film most closely. 

It is more important with sound recordings than with silent pic- 
tures to be able to make .a reasonably long run without stopping to 
rewind, since discontinuities in sound are more serious. To obtain 
the maximum length of run, the camera must take as little power as 
possible and the driving spring be as large as weight and space will 
permit. The present design provides for running 25 feet at one 
winding. Although a greater length might be considered desirable, 




FIG. 9. 



Complete amateur sound picture recording 
equipment. 



this would mean added weight and bulk, and amateur picture pro- 
ducers would wish, with few exceptions, to make their shots short. 
A definite stop is provided so that the camera comes to an abrupt 
stop rather than being permitted to run when the spring is so far 
unwound that the speed might fall and the sound quality be ruined. 
Careful studies of battery life were carried out to make sure that 
the small cells, which are necessary to keep the weight down, would 
give the required service without too much drop in voltage for main- 
taining amplifier output or, what is still more critical, for maintaining 
the exposing lamp within proper limits of brightness. The battery 
box carries duplicate batteries, which may be alternated when mak- 
ing pictures of considerable length. The switching arrangement in 



88 E. W. KELLOGG [J. s. M. p. E. 

the camera is such that the exposing lamp is lighted only while the 
camera is running or while the shield is held aside for monitoring. 
The operating button may be locked in the running position if de- 
sired. This permits the operator to center his attention upon other 
matters more completely, or to step into the picture. 

The finder is like a telescope in form but employs a concave ob- 
jective and a magnifying eyepiece. This gives an image of reduced 
size, but very clear and bright. The finder is placed as close to the 
photographic lens as possible to minimize parallax. A parallax ad- 
justment is provided for use when the subject is less than five feet 
from the camera. 




FIG. 10. Sixteen-mm. sound camera equipment set 
up on tripod. 

A three-lens turret is provided. This adds little to the cost of the 
camera if only the general-service universal-focus lens is wanted, 
but leaves open the possibility of special lenses for high-speed or 
telephoto work. Three rectangles are engraved in the finder, which 
show the fields included with the standard 1-, 2-, and 4-inch focus 
lens, respectively. In order that adjustable-focus lenses may be 
set to give maximum sharpness, a place is provided where a critical 
focusing device may be added. The turret is turned so as to bring 



Jan., 1935] 16-MM. SOUND MOTION PICTURES 89 

the lens to be adjusted opposite the focusing position, where the 
image can be viewed through a magnifying eyepiece. The turret is 
then turned back and the lens will be in accurate focus on the film. 

All sound records must be made at the standard speed of 24 frames 
per second, but in the case of silent shots some saving of film results 
from operation at lower speed, and the camera is therefore designed 
with a two-speed adjustment permitting operation at either 16 or 24 
frames a second. 

Fig. 9 shows the complete equipment, with microphone and amp- 
lifier. If maximum mobility is wanted the amplifier and battery 
boxes can be carried on a belt and the camera in the hands. For 
general purposes it is more satisfactory to mount the amplifier and 
battery box on the tripod with the camera, as shown in Fig. 10. 

ACKNOWLEDGMENT 

The developments described in this paper have engaged the 
thoughts of many of our engineers and members of the sales staff, and 
the desirability of various features has been checked by extended 
trials. While it is impossible to give appropriate credit to all who 
have contributed, mention should be made of the contributions of 
A. C. Blaney, in his studies of film emulsions and processing; of 
A. Shoup, in working out the efficient and compact amplifiers; of R. P. 
May, R. L. Hanson, H. C. Holden, and B. L. Hubbard in the pro- 
jector development; G. L. Dimmick and L. T. Sachtleben in working 
out many of the optical and related problems; C. N. Batsel and I. J. 
Larson in devising the mechanical features of the camera; and A. G. 
Zimmerman in adapting its design to manufacture. 

REFERENCES 

1 BATSEL, C. N., AND BAKER, J. O.: "Sound Recording and Reproducing 
Using 16-Mm. Film," /. Soc. Mot. Pict. Eng., XXI (Aug., 1933), No. 2, p. 161. 

2 COOK, E. D.: "The Aperture Effect," /. Soc. Mot. Pict. Eng., XIV (June, 
1930), No. 6, p. 650. 

3 STRYKER, N. R. : "Scanning Losses in Reproduction," /. Soc. Mot. Pict. Eng., 
XV (Nov., 1930), No. 5, p. 610. 

4 DIMMICK, G. L.: "High-Frequency Response from Variable- Width Records 
as Affected by Exposure and Development," /. Soc. Mot. Pict. Eng., XVII (Nov., 
1931), No. 5, p. 766. 

'DIMMICK, G. L.: "Galvanometers for Variable-Area Recording," /. Soc. 
Mot. Pict. Eng., XV (Oct., 1930), No. 4, p. 428. 



90 E. W. KELLOGG 

6 BATSEL, C. N.: "A Non-Slip Sound Printer," J. Soc. Mot. Pict. Eng., XXIII 
(Aug., 1934), No. 2, p. 100. 

7 DIMMICK, G. L., BATSEL, C. N., AND SACHTLEBEN, L. T.: "Optical Reduction 
Sound Printing," /. Soc. Mot. Pcit. Eng., XXIII (Aug., 1934), No. 2, p. 108. 

8 BAKER, J. O.: "Sixteen-Mm. Sound on Film," /. Soc. Mot. Pict. Eng., XXII 
(Feb., 1934), No. 2, p. 139. 

9 MAY, R. P.: "16-Mm. Sound Film Dimensions," J. Soc. Mot. Pict. Eng., 
XVIII (April, 1932), No. 4, p. 488. 

10 BATSEL, C. N., SACHTLEBEN, L. T., AND DIMMICK, G. L.: "A 16-Mm. 
Sound Recording Camera," /. Soc. Mot. Pict. Eng., XXIII (Aug., 1934), No. 2, 
p. 87. 



SOCIETY ANNOUNCEMENTS 

PACIFIC COAST SECTION 

The Annual Election Dinner of the Section was held on December 13, fol- 
lowed by the announcement of the successful candidates for Section offices for 
1935. The results of the election were as follows : 

Chairman: G. F. Rackett 
Sec.-Treas.: H. W. Moyse 
Manager: K. F. Morgan 

The second Manager of the Section is W. C. Harcus, whose term expires Dec. 
31, 1935. Mr. E. Huse remains a member of the Board of Managers as Past- 
Chairman, hi addition to holding the post of Executive Vice-President of the 
Society for 1935. 

Following the elections and other business, the prize-winning 16-mm. films 
selected by the American Society of Cinematographers for 1932, 1933, and 1934 
were shown, followed by a showing of unusual 8-mm. films. 

ATLANTIC COAST SECTION 

The usual monthly meeting was held at the Hotel Pennsylvania, New York, 
on December 19, at which time Mr. W. B. Rayton presented a tutorial paper on 
the subject of "Lens Design/' The paper was non-mathematical, its purpose 
being to present the subject hi a manner understandable by those whose profes- 
sion is not that of optics, although it contained much information of value to 
experts. 

The results of the annual election of officers of the Section for 1935 were as 
follows: 

Chairman: L. W. Davee 

Sec.-Treas.: D. E. Hyndman (reflected) 

Manager: H. Griffin 

The second Manager of the Section is M. C. Batsel, whose term expires Dec. 31, 
1935. Mr. H. G. Tasker remains a member of the Board of Managers as Past- 
Chairman, in addition to holding the post of President of the Society for 1935. 

BOARD OF GOVERNORS 

The next meeting of the Board will be held on January 11, at the Hotel 
Pennsylvania, New York, N. Y., at which time the operating budget of the 
Society for 1935 will be framed, and further arrangements for the Hollywood Con- 
vention, to be held on May 20-24, incl., will be made. The new officers of the 
Society for 1935 will officiate at that meeting for the first time, and arrangements 
will go forward toward the establishment of the various technical and other 
committees, and the Sectional Committee on Motion Picture Standardization 
according to the procedure of the American Standards Association. 

91 



92 SOCIETY ANNOUNCEMENTS 

SPRING, 1935, CONVENTION 

As announced previously, the Spring Convention will be held this year at 
Hollywood, May 20 to 24, inclusive, with headquarters at the Hotel Roosevelt. 
Members of the Society are urged to make every effort to attend, and contribute 
to making this Convention the greatest and most interesting in the history of the 
Society. The Pacific Coast Section Board of Managers, under the Chairmanship 
of Mr. E. Huse, is collaborating with Mr. W. C. Kunzmann, Convention Vice- 
President, in arranging the details of the Convention; and the Papers Committee, 
directed by Mr. J. I. Crabtree, Editorial Vice-President, and Mr. J. O. Baker, 
Chairman, promises a most outstanding program of technical papers and demon- 
strations. 

Those members who do not reside in the West are urged to arrange their vaca- 
tions to coincide with the Convention, in case their business arrangements do not 
otherwise permit their attendance. Advantage can be taken of the special tour- 
ists' railroad rates effective May 15. 

PROJECTION PRACTICE COMMITTEE 

At a meeting held at the Paramount Building, New York, N. Y., December 
12, plans were laid for the work of the Committee during the coming year. 
Among the various subjects to be considered are the possible recommendations 
for projection screen brightness, and simple and practical methods and instru- 
ments for measuring it; and a complete re-presentation of the projection room 
layouts and maintenance technic presented in the 1931 report of the Committee, 
bringing it up to date and in accordance with the latest and most acceptable 
practice in connection therewith. The earlier report, published in the Aug., 
1931, issue of the JOURNAL, has been widely used throughout the industry, and its 
growing importance has justified and made requisite some revisions in accordance 
with the changing needs and practices. 

STANDARDS COMMITTEE 

At a meeting held at New York on December 14, subjects for study and 
possible standardization in 1935 were considered, among which were: screen 
brightness; sound sprockets; edge-guiding of film in cameras, printers, and 
projectors; and camera mechanisms. 



JOURNAL 



OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 

Volume XXIV FEBRUARY, 1935 Number 2 



CONTENTS 

Page 

Some Characteristics of 16-Mm. Sound by Optical Reduction 
and Re-Recording . . . . C. N. B ATSEL AND L. T. SACHTLEBEN 95 

Background Projection for Process Cinematography 

G. G. POPOVICI 102 

High-Intensity Mercury and Sodium Arc Lamps 

L. J. BUTTOLPH 110 

Piezoelectric Loud Speakers A. L. WILLIAMS 121 

Reflecting Surfaces of Aluminum J. D. EDWARDS 126 

Simple Theory of the Three-Electrode Vacuum Tube 

H. A. PIDGEON 133 

The 16-Mm. Sound-Film Outlook W. B. COOK 175 

A Physical Densitometer for Sound Processing Laboratories . . 

F. L. EICH 180 

Society Announcements 184 



JOURNAL 

OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 



SYLVAN HARRIS, EDITOR 

Board of Editors 
J. I. CRABTREE, Chairman 

O. M. GLUNT A. C. HARDY L. A. JONES 

J. O. BAKER 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1935, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 



Officers of the Society 

President: HOMER G. TASKER, 4139 38th St., Long Island City, N. Y. 
Past-President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
Executive Vice-P resident: EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, 

Calif. 

Engineering Vice-P resident: LOYD A. JONES, Kodak Park, Rochester, N. Y. 
Editorial Vice-President: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Financial Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Convention Vice-President: WILLIAM C. KUNZMANN, Box 6087, Cleveland, Ohio. 
Secretary: JOHN H. KURLANDER, 2 Clearfield Ave., Boomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York, N. Y. 

Governors 

MAX C. BATSEL, Front & Market Sts., Camden, N. J. 
LAWRENCE W. DAVEE, 250 W. 57th St., New York, N. Y. 
ARTHUR S. DICKINSON, 28 W. 44th St., New York, N. Y. 
HERBERT GRIFFIN, 90 Gold St., New York, N. Y. 
WILBUR B. RAYTON, 635 St. Paul St., Rochester, N. Y. 
SIDNEY K. WOLF, 250 W. 57th St., New York, N. Y. 



SOME CHARACTERISTICS OF 16-MM. SOUND BY 
OPTICAL REDUCTION AND RE-RECORDING* 

C. N. BATSEL AND L. T. SACHTLEBEN** 

Summary. Some characteristics of 16 -mm. sound by optical reduction and re- 
recording from 35-mm. film are discussed: by means of loss curves characteristic of the 
two methods; by reviewing the effects upon the sound quality of compensation for such 
losses; and by noting the effects, other than simple output loss, that arise in some of 
the steps of each process. 

Sixteen-millimeter sound is obtainable from 35-mm. film by optical 
reduction or by re-recording. Many variations of the optical reduc- 
tion process are possible, since it is wholly an optical-photographic 
process; but in re-recording it is necessary to work always from a 
print so prepared that distortions of a photographic nature are elimi- 
nated as far as possible, because distortions of a photographic origin 
can not be compensated for in any way once they have been impressed 
upon an electrical circuit. In this paper two variations of the opti- 
cal reduction process are discussed in relation to the standard re- 
recording process. 

The apparatus used in optical reduction printing is an anamor- 
phote optical system imaging the 35-mm. sound-track upon the 
16-mm. film in accordance with S. M. P. E. standards, and is described 
in a previous paper. 1 The films are moved uniformly past the scan- 
ning and printing points by a suitable mechanical arrangement that 
insures against speed variations. When working from an original 
negative to a final positive print no additional apparatus is required ; 
but when working from a positive, a 35-mm. contact sound printer 
and a 16-mm. contact sound printer are required for producing the 
original positive and the final 16-mm. positive after the 16-mm. nega- 
tive has been produced by optical reduction. Both these variations 
of optical reduction are discussed in this paper. 

In re-recording, the apparatus required comprises a 35-mm. contact 
sound printer for preparing the print, a 35-mm. reproducing head 

* Presented at the Fall, 1934, Convention at New York, N. Y. 
** RCA Victor Co., Camden, N. J. 

95 



96 



C. N. BATSEL AND L. T. SACHTLEBEN [J. S. M. p. E. 



with optical system, an amplifier, a 16-mm. recording head with 
optical system, and a 16-mm. contact sound printer for preparing the 
16-mm. print. 




FREQUENCY IN CYCLES PEN SECOND 

FIG. 1. Frequency characteristics of various steps in re-recording 35-mm. sound 

on 16-mm. film. 




FIG. 2. Over-all compensation required for constant output from 16-mm. prints 

by re-recording. 



Feb., 1935] CHARACTERISTICS OF 16-MM. SOUND 97 

The data presented here were obtained during the course of de- 
velopmental work on both optical reduction printing and re-recording 
apparatus from frequency films prepared to show the performance 
of the two systems, and care was always taken so to expose and proc- 
ess the sound-tracks that distortion was kept at a minimum. 2 All 
recording and reproducing equipment involved in this work, both 
16-mm. and 35-mm., used V2-mil recording and reproducing slits, 
and the data are therefore based upon these values. Only the vari- 
able-width system was employed. 

Considering first the re-recording process, the sound is electrically 
reproduced from a properly prepared print of the original 35-mm. 
negative, amplified, and fed into a 16-mm. recording system, to pro- 

Step Accrued loss at 5000 cycles 

I 35-mm. Negative 4.3 db. 39% 



35-mm. Contact Print 3.3 32 



35-mm. Reproducer 3.5 33 



16-mm. Negative 



16-mm. Contact Print 



21.6 92 



16-mm. Reproducer 23.6 92.5 



FIG. 3. Steps and losses in re-recording 35-mm. on 16-mm. 

duce a 16-mm. negative from which the final properly prepared 16-mm. 
print is obtained. Six definite steps with associated losses are 
involved before reaching the loud speaker of the 16-mm. reproducer. 
Losses occur in the 35-mm. negative due to the finite width of the 
recorder slit and due to certain film characteristics. A slight gain 
occurs in making the 35-mm. print. Correcting for reproducer slit 
loss, the total loss incurred in recording and producing a print of the 
35-mm. negative is determined by subtracting curves 1 and 2 of Fig. 1, 
which are the response curves of a 16-mm. print of the S. M. P. E. 
test print by re-recording. At 5000 cycles, this loss amounts to 
about 3.3 db. A further loss results from scanning the print with a 
finite slit, bringing the total loss to about 3.5 db. This same series of 



98 



C. N. BATSEL AND L. T. SACHTLEBEN [J. S. M. p. E. 



losses is repeated, wavelength for wavelength, in recording, printing, 
and reproducing the 16-mm. sound record, resulting in the response 
curve 4 of Fig. 1, which shows the over-all loss, and therefore the over- 




rncaucNcv m CVCLCC ft* SECOND 



FIG. 4. Frequency characteristics of various steps in optically reducing 35-mm. 
sound to 16-mm., working from the negative 




FIG. 5. Over-all compensation required for constant output from 16-mm. print 
by optical reduction from either the 35-mm. negative or the print. 



Feb., 1935 J CHARACTERISTICS OF 16-MM. SOUND 99 

all requisite compensation at 5000 cycles, to be 23.6 db. Fig. 2 is 
the over-all compensation curve required in re-recording, and Fig. 3 
is a block diagram of the steps involved in re-recording and the ac- 
crued losses at the end of each step, for a frequency of 5000 cycles. 
When printing the 16-mm. sound-track by optical reduction from 
the original 35-mm. negative, using the anamorphote optical system, 
only three loss-producing steps are involved. The 35-mm. negative 
is prepared as for re-recording, but no steps involving reproducing or 
recording losses intervene between it and the final 16-mm. print, be- 
cause it is produced by optically imaging the 35-mm. negative upon 
the 16-mm. emulsion in the ratio of the normal speeds of the two 
films as they pass uniformly through the printer. (In the transverse 
plane, the 35-mm. negative is optically imaged upon the 16-mm. 
emulsion in the ratio of the standard widths of the two sound-tracks.) 

Step Accrued loss at 5000 cycles 

35-mm. Negative 4.3 db. 39% 



16-mm. Optical Reduction Print 10.6 70.5 



16-mm. Reproducer 12.6 76.5 



FIG. 6. Steps and losses in optical reduction printing 35-mm. to 16-mm., working 

from the negative. 

Fig. 4 is the response curve of an optical reduction print of the 
S. M. P. E. test negative, and shows an over-all loss of 12.6 db. at 5000 
cycles, including the final reproducer scanning loss. Fig. 5 is the 
curve of over-all compensation required in optical reduction printing 
from a 35-mm. negative, and Fig. 6 is a block diagram of the steps 
involved and the accrued losses at the end of each step, at 5000 cycles. 

The superiority of 16-mm. prints made by this process lies not only 
in the improvement in the frequency response indicated by these re- 
sults, but in eliminating the contact printing steps that always intro- 
duce a certain "fuzziness" into the print as a result of slippage and 
poor contact between the negative and the raw positive stock. 

The optical reduction process may be varied so as to work from a 
35-mm. print rather than from a negative, in which case a 16-mm. 
negative is first obtained by optical reduction, from which a 16-mm. 
print is subsequently made by contact printing. Five steps, intro- 
ducing losses, are involved before reaching the reproducing loud 



100 



C. N. BATSEL AND L. T. SACHTLEBEN [J. S. M. P. E. 



speaker. By examining Fig. 7 it is seen that the final response of the 
16-mm. copy of the S. M. P. E. test print does not differ appreciably 
from that of the optically reduced print of the S. M. P. E. test negative. 




FREQUENCY IN CYCLES PER SECOND 



FIG. 7. Frequency characteristics of various steps in optically reducing 35-mm. 
sound to 16-mm., working from the print. 



Step 



35-mm. Negative 



Accrued loss at 5000 cycles 

4.3 db. 39% 



35-mm. Contact Print 



16-mm. Optical Reduction Neg. 



3.3 



10.7 



32 



71 



16-mm. Contact Print 



11.1 



72 



16-mm. Reproducer 



13.1 



78 



FIG. 8. Steps and losses in optical reduction printing 35-mm. to 16-mm., working 

from the print. 

This should be expected, since the first step, that of contact printing 
the 35-mm. negative, results in a gain in response over the negative, 
which probably offsets subsequent losses in printing the 16-mm. opti- 
cally reduced negative, the wavelengths of which are much 



Feb., 1935] CHARACTERISTICS OF 16-MM. SOUND 101 

shorter. It is obvious that the over-all compensation required for 
either variation of the optical reduction printing process is the same. 
Fig. 8 is a block diagram of the steps involved in optical reduction 
printing from a 35-mm. print, and the accrued losses at the end of 
each step, both in decibels and in per cent, at 5000 cycles. 

To summarize, an examination of the characteristics of 16-mm. 
sound by optical reduction and re-recording shows that the compari- 
son stands in favor of optical reduction, as producing a better 16-mm. 
sound record when no compensation is introduced, and as requiring 
vastly less compensation when the aim is to produce a 16-mm. print 
having a flat response curve, the difference at 5000 cycles being ap- 
proximately 12 db., which represents a compensation ratio of 4 to 1. 
Thus, in producing 16-mm. sound records flat to 5000 cycles from 35- 
mm. film, the optical reduction process permits the average recorded 
level of the 35-mm. negative to be greater than that permitted by the 
re-recording method, depending upon the frequency content of the 
sound. For this reason, the ratio of distortion introduced by com- 
pensating for the higher frequencies to the average signal level will be 
less in optical reduction than in re-recording. 

The gain that optical reduction shows over re-recording lies mostly 
in eliminating the 16-mm. recording loss, which at 5000 cycles 
amounts to about 12 db. (equal to the loss of the 35-mm. film at 
12,500 cycles). It should be noted that this gain can not be achieved 
except when the optical reduction is in the true ratio of the speeds of 
the two films, as in the case of the anamorphote system or a system 
that reduces the track width also in the same ratio. Reduction op- 
tical systems that work near unit magnification and employ fine slits 
to compensate for the difference of film speed involve in greater or 
lesser degree the losses inherent in the recording phase of re-recording. 
This may be readily appreciated when it is remembered that in both 
cases the 16-mm. sound record is produced by exposing a moving 
emulsion to a slit image of varying length. In so far as contact 
printing may be used in the two processes, the effect is about the same. 

REFERENCES 

GIMMICK, G. L., BATSEL, C. N., and SACHTLEBEN, L. T.: "Optical Reduction 
Sound Printing," /. Soc. Mot. Pict. Eng., XXIII (Aug., 1934), No. 2, p. 108. 

2 SANDViK, O., HALL, V. C., and STREIFFERT, J. G.: "Wave Form Analysis of 
Variable-Width Sound Records," /. Soc. Mot. Pict. Eng., XXI (Oct., 1933), 
No. 4, p. 323. 



BACKGROUND PROJECTION FOR PROCESS 
CINEMATOGRAPHY* 



G. G. POPOVICI** 

Summary. A process of background projection is described employing a side- 
guided projector, and a cut-out in the gate permitting images of the edge of the run- 
ning film and two film perforations to be projected upon the screen simultaneously 
with the image of the picture, so as to check the steadiness of the picture. Screen 
illumination requirements, effects of film shrinkage, and pilot-pin registration are 
discussed. 

Background projection for process photography now used exten- 
sively in all motion picture studios is more economical than the Dun- 
ning process, the color-separation method, the use of printing masks, 
and other arrangements. It is probably the simplest process that 
has yet been devised, and with proper care can produce satisfactory 
results. It consists in projecting a picture of the desired background, 
which has been photographed previously, upon a translucent screen, 
the foreground action of the picture taking place in front of the screen. 
The foreground objects are so lighted as to balance the screen illumina- 
tion, and the whole is then finally photographed as a composite scene. 

The most important factor involved in the background projection 
process is steadiness of the picture. To begin with, the negative 
film used for making the "plate" or "key," which are the names given 
to the scene to be projected, has to be very accurately perforated, and 
only fresh stock must be used. Eastman and Du Pont have both de- 
veloped such an accurately perforated film, having a finer grain with 
a slightly lighter contrast than the regular negative emulsions, and 
having approximately the same camera characteristics as the regular 
Super-Sensitive Panchromatic negative stock. 

The camera used for shooting the plates must produce absolutely 
steady pictures. Its ability to do so depends upon the mechanism 
of the camera itself and the manner in which the camera operates. 
A standard Mitchell camera having a shutter opening of 170 degrees 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 
** Eastern Service Studios, Inc., Long Island City, N. Y. 
102 



BACKGROUND PROJECTION 



103 



is used for making the plates, and its steadiness is carefully checked 
periodically to assure good results. 

The prints to be projected present a problem that has not as yet 
been solved satisfactorily. Due to pilot-pin registration in step 
printing, shrinkage of the developed negative causes some difficulty. 
If the pin is designed for the sprocket-hole of fresh positive stock, 
using the Bell & Howell perforation, during printing the negative 
sprocket hole will be forced down upon the pin by a slight movement of 
the finished positive due to shrinkage. No matter how the pin is de- 
signed, so long as shrinkage is to be considered some difficulty will 
always exist. Only unshrinkable stock would be completely satis- 
factory for pilot-pin registration. 




FIG. 1. 



Set-up of projector, camera, translucent screen, lights, and 
actors, for rear projection process photography. 



Projection of the plate positive has successfully been accomplished 
in our studios. Contrary to West Coast procedure, we use a side- 
guided projector, designed by the International Projector Corp. 
To assure perfect steadiness, the gate and film-trap have been espe- 
cially designed. The steel guide is hard and has been ground. The 
tension springs are carefully balanced to assure perfect side-guiding 
and proper gate pressure. The Geneva intermittent movement is 
selected for its precision, having no tolerances, and the intermittent 
sprocket is of the type recently approved by the Society and has also 
no tolerances. 

An ingenious device incorporated in this projector enables the 
projectionist to check at any time the steadiness of the projected 



104 



G. G. POPOVICI 



[J. S. M. P. E. 



picture. Unsteady projection may be attributed to the print or to 
the projector. To be able to check one or the other at once, a cut-out 
is provided in the gate permitting an image of the edge of the running 
film and two sprocket holes to be projected upon the screen with the 
picture. If the image of the sprocket holes is steady and the picture 
image is unsteady, the print is obviously the cause of the unsteadiness. 
If the sprocket hole image is unsteady, there still remains the ques- 
tion whether the projector is working properly, because instances 
have been found in which the film had shrunk unevenly, causing the 
sprocket holes to be unsteady. Such a test should always be made 
with fresh stock. (It is of vital importance that the projector al- 




FIG. 2. Projector gate dismantled, showing 
aperture through which the image of the film 
perforations is projected. 



ways be kept clean, because an accumulation of dirt may cause un- 
steadiness of the projected image.) 

The advantages of this projector over those employing the claw 
movements are outstanding. This projector has a shutter opening 
of 270 degrees, as compared with 180 degrees for the claw movement 
type. The picture can be framed, whereas with the other type it 
cannot; the intermittent works cold : only the film is exposed to the 
heat, whereas in the claw type machine the claw and pins are continu- 
ously exposed to the heat, and subject to expansion. 

The lamp is a Hall & Connolly HC-10, of the thermostatically con- 
trolled type, using 13.6-mm. positive and 11-mm. metal-coated nega- 



Feb., 1935] BACKGROUND PROJECTION 105 

tive carbons. The current is 165 to 200 amperes. The condensers 
are standard 5127 for the rear, and 5128 for the front element. The 
position of the condenser lenses with respect to the arc and projector 
differs from that in ordinary theater projection, owing to the neces- 
sity of eliminating the hot spot which ordinarily would appear upon 
the screen after the rephotographing. 

The projection lenses are the Apermax-Cooke 5Vi-inch//1.9, and 
the 6Y2-mch//2.3. These lenses proved to be superior to any other 
lenses tested so far. 

In the beginning a Trans-Lux screen was used, having very good 
diffusion but low transmission, thereby handicapping scenes requiring 
big picture sizes. However, it proved very satisfactory so far as 
the hot spot and satisfactory viewing angle were concerned. Later 
a rear projection screen made by Fox on the West Coast was obtained, 




FIG. 3. Projector gate assembled. 

having nearly double the transmission of the Trans-Lux screen but 
a narrower satisfactory viewing angle. 

Photographing the plate requires experience and skill on the part 
of the cameraman in charge. He must visualize in advance what 
the projected image should be, and looking through the finder of the 
camera when photographing the plate he must take into considera- 
tion the foreground that will be in front of the screen when making 
the composite scene. Moreover, he must judge accurately the ex- 
posure required for a negative density contrast of 0.80 to 0.95 after 
the film is normally developed. 

When buildings or other subjects that must appear steady on the 
screen are photographed, the camera must stand very steadily upon 
its tripod, which may have to be tied down. When photographing 



106 G. G. POPOVICI [J. S. M. P. E. 

from bridges, long focal length lenses must be avoided because of the 
vibration of the bridge. Wind also causes unsteady negatives, and 
it is suggested that plates be not photographed on windy days. 

From the finished negative plate three positives are made: one 
called "normal," another one printer-light higher, and the third one 
printer-light lower. From these prints the one best suited to the 
composite scene is chosen. These prints are special in so far as they are 
made on a selected step printer. For lining up the scene an ordinary 
print made on a continuous-running Bell & Howell printer is used. 
Each work print is preceded by 15 feet of a sharp criss-cross chart to 
permit focusing the picture accurately. 

The actual operation is as follows: Required sets and props are 
built and erected before the translucent screen. Necessary lighting 
equipment is approximately placed in position. Plenty of gobos are 
on hand to shade the screen from light striking the foreground. A 
projected picture of the approximately correct size is thrown upon the 
screen. If the size is not satisfactory, lenses are changed or the pro- 
jection machine is moved to attain the desired result. Then the 
lighting is balanced by measuring first the light striking the fore- 
ground, and then the intensity of the projected light. The measure- 
ments are made with the Weston photronic meter. 

A final photographic test is made as follows : The foreground is il- 
luminated, an interlocking phase is switched on, and the camera 
shutter is set to a standard position. Then the clutch of the pro- 
jector is closed in the corresponding shutter-synchronizing position, 
and the projector and camera are operated synchronously for about 
30 feet of film. From the end of this test-exposure two feet are de- 
veloped and fixed by hand, and then enlarged on paper so as to show 
what changes might be required. After obtaining a good test, the 
scene is finally photographed and recorded. 

In conclusion, the author expresses his sincere appreciation to 
Messrs. L. W. Davee and H. Griffin who cooperated in making this 
background projection process a success. 

DISCUSSION 

CHAIRMAN CRABTREE: In the scene of Riverside Drive, where the girls were 
seated on the parapet and the boys were going through their antics in front of 
them, the girls didn't seem to be in the least interested in what the boys were 
doing. Obviously, if there are any animate objects in the picture when the back- 
ground scene is shot they ought to be doing what you intend them to be doing 
in the composite picture. 



Feb., 1935] BACKGROUND PROJECTION 107 

How are the screens made? Originally ground glass was used, but I under- 
stand that most screens consist of fabric sprayed with, say, cellulose acetate or 
some similar transparent material. 

MR. POPOVICI : I am sorry I can not give you any definite information on cel- 
lulose screens, because we used only the Fox screen, which, so far as I know, is 
made by spraying certain materials upon a flat plate of glass, and then peeled 
and processed in a secret way. As regards the Trans-Lux screen, which was 
used on the shots you saw today, it is silk impregnated with gelatin with a little 
color pigment in it. The gelatin is pressed against a matte surface in such a way 
as to provide a high diffusion factor. The transmission of the screen is, however, 
very low; the one we have used so far transmits only 17 per cent of the light. 
The viewing angle decay is very, very small, even enabling us to make shots 
when running with a camera from the side, or in and out, without difficulty or 
without any difference of density in the re-photographed image. 

On the Fox screen, however, the slightest angle that is introduced between the 
camera and the projection axis will cause trouble. It is a high-transmission 
screen, however, transmitting 30 per cent of the light, and enabling one to bal- 
ance out the front lighting very easily in order to attain a good over-all balance 
of lighting. The "hot spot" is a little more noticeable on a Fox screen than on a 
Trans-Lux. That is why we place the projection machine as far as we can 
from the screen. 

MR. MITCHELL: You said that you use the condensers to eliminate the hot 
spot. Can you give us any details? 

MR. POPOVICI: It has been a very empirical procedure. The Bausch & 
Lomb specifications require that the condenser be set about 14 inches from the 
projection machine aperture and the distance from the carbon to the condenser 
lens about 8 inches. I arrived at my settings, as a matter of fact, after much 
experimenting, measuring the light on the other side of the screen with a Macbeth 
illuminometer and a Weston photronic meter. I found a large difference be- 
tween readings made with Bausch & Lomb specifications and our own. The 
distance between the condenser lenses remains unchanged. The lamp house is 
pushed back 2lVz inches, and the distance between the condenser lens and the 
carbon is only 3 inches, instead of 8. 

The projected concentrated spot from the condenser lenses on the aperture 
is quite characteristic in that a concentrated spot of great intensity covers the aper- 
ture, and from that point on, the remainder of the circle, which is not used, is of 
lower intensity. We try to get collimated light, which would be ideal for that 
kind of process, but so far we haven't been able to get it with sufficient intensity 
for all requirements. 

MR. FRANK: In most instances are the original outdoor scenes taken specifi- 
cally for the picture, or do you have a library? Are sound effects ever recorded 
with the original outdoor scene, or are they always dubbed separately onto the 
final film? 

MR. POPOVICI: There is, so to speak, a small library for outdoor scenes, but 
we do not use it for shots that have to be steady. We would rather send out a 
cameraman to shoot a new scene, because the negative has lain quite some time 
in the vault and has shrunk, and it is difficult to print a shrunken negative. It 
is cheaper to make the shot anew. For scenes of the kind required for taxicab 



108 G. G. POPOVICI [j. s. M. p. E. 

shots or running automobiles or trains, where steadiness is not required, we use 
library shots, of course. Synchronizing the sound effect is usually done in dub- 
bing. 

MR. J. CRABTREE: I believe those are Bell & Howell perforations on the 
negative and square perforations on the print. Have you used the same kind 
of perforations on both negative and print? 

MR. POPOVICI: We use exclusively the Bell & Howell perforation all the way 
through. 

MR. J. CRABTREE: Is it preferable to the square perforation? 

MR. POPOVICI: It is, because the print has to be very steady, and running 
Bell & Howell perforations against Eastman perforations would be quite a 
difficult task. 

MR. J. CRABTREE: But the Society has recommended the adoption of the 
square kind of perforation throughout. 

CHAIRMAN CRABTREE : The question is why would not the square perforation 
with the square pilot pin be just as good as the Bell & Howell pilot pin with the 
Bell & Howell perforation? 

MR. POPOVICI: It would be, so far as pertains to printing, but in printing 
there are two sprocket holes to accommodate. The negative has shrunk a small 
amount, and the positive fresh stock has not shrunk. If you take a positive 
stock that will fit snugly upon a pilot pin, this undeveloped stock will be too 
big for the hole of the negative that has been developed and shrunk, and you will 
have to force the negative over the pilot pin. 

MR. J. CRABTREE : That is true irrespective of the type of perforations. Why 
not measure the pitch of the negative, and perforate the material on which you 
are going to print it to the pitch of the negative. 

MR. POPOVICI: That would be one way out. We are designing a printer that 
has no pilot pin registration. It has only one up-and-down registration to make 
sure that both holes are registered in the up-and-down way. 

MR. J. CRABTREE: My question was put simply to determine whether the 
recommendations of the Society, since they have been adopted, are causing any 
difficulty in using the square perforations. 

MR. POPOVICI: I have nothing against it. As a matter of fact, it would be 
all right to use only one kind of perforation, but the difficulty lies in the existing 
camera equipment using the Bell & Howell perforation, which is scattered all 
over the world. 

CHAIRMAN CRABTREE: As Mr. Popovici says, in view of the fact that there 
are a large number of library shots having the Bell & Howell perforation, ap- 
parently the producers prefer to standardize on the Bell & Howell perforation, 
unless someone can establish practical advantages for the square perforation. 

MR. RICKER: We have found it necessary to use the Bell & Howell perforation 
in similar work with the positive. 

CHAIRMAN CRABTREE: Why? 

MR. RICKER: Because we got a steadier picture. 

CHAIRMAN CRABTREE: Did you try a special pin to fit the square perforation, 
or did you use the old Bell & Howell registration pin? 

MR. RICKER: I presume we used the old Bell & Howell pin. 

CHAIRMAN CRABTREE: Obviously, it wouldn't work. 



Feb., 1935] BACKGROUND PROJECTION 109 

MR. RICKER: I want to add one word of caution. Sometimes a cameraman 
or laboratory man will have occasion to rewind film. Since in the film 
factory the film is undoubtedly edge-guided during perforation, after slitting 
the film from an edge-guided side, you must be careful to keep that same 
edge-guide inside, both in the camera and in the projector, because you will 
find when it is edge-guided for perforation on one side, reversing the film and 
projecting it guided on the other side will result in a variation in the position of 
the perforations. 

MR. POPOVICI: That is true. As a matter of fact, we know which side is 
guided during perforation and we always try to keep the same side guided all the 
way through. 

CHAIRMAN CRABTREE: Has anyone made registration pins specifically for 
the square perforation, and compared them for steadiness against the Bell & 
Howell pin with the Bell & Howell perforation? 

MR. MITCHELL: We haven't made really exhaustive tests hi that way, but 
we studied the matter a lot when the rectangular perforation was first proposed. 
The Bell & Howell perforation, of course, has curved sides. If there is any 
shrinkage, the pilot pin of the camera or the driving tooth of the printer sprocket 
fits in the way of a taper fit. In other words, if there is any shrinkage of a reason- 
able amount, it is to some extent self -compensating. 

Hollywood prefers the pilot pin for registration. There has been quite a 
tendency to use oversize pilot pins. We have been called upon quite often to 
make pilot pins that would be exactly 0.0001 inch greater than the perforation 
size, the feeling being that the film is jammed down upon the pin and that the 
film is sufficiently elastic to stretch during registration and shrink back to its 
original shape without deformation. Within certain limits I believe that that 
characteristic of the film would apply also with shrinkage, and that the stretching 
characteristic of the film would accommodate at least a small percentage of 
shrinkage. 

MR. POPOVICI: Oversized pins are very good in optical printing where you 
make two or three pictures a second. But when the movement is fast, you will 
agree that the sprocket hole is definitely deformed when it goes over the pilot 
pin. Disregarding printing, and considering now projection, the projection 
machines on the coast are all equipped with pilot pins. The pins start to work 
in the cold state; when 200 or 300 feet of film are shot on a screen, the pilot pin 
becomes hot and expansion occurs. The film has a tendency to shrink from the 
heat; the pilot pin swells. What results? A steady shot in the beginning, but 
movement the third time it is shot. 

MR. MITCHELL: I always thought the machine was warmed up before start- 
ing. 

MR. POPOVICI : Perhaps ; but there are two factors working against each other. 
In side-guiding the film isn't forced at all ; the only pressure that acts comes from 
the springs that hold the film against the gate. Three sprocket teeth, instead 
of one pin, engage the film; hence the torque of the sprocket drum against the 
film is so much less than on a pilot pin movement. However, the motion is 
twice as fast, which balances it out, and still it is a little less than with the pilot 
pin. 



HIGH-INTENSITY MERCURY AND SODIUM ARC LAMPS 



L. J. BUTTOLPH** 

Summary. Characteristics of the sodium and high-intensity mercury arcs of 
interest to motion picture engineers are discussed, including starting, stability, regu- 
lation, power-factor, and multiple and series operation. Energy distribution as 
related to the sensitivity curves of photographic emulsions and to the use of filters is 
shown by graphs. Methods of controlling the light distribution and intensity are 
explained. 

The pioneer service of the Cooper Hewitt lamp in the motion pic- 
ture field encourages thought of further service from the latest de- 
velopments in gaseous conduction arcs. The new high-intensity 
mercury and sodium lamps differ from other Cooper Hewitt lamps 
in their relative compactness and increased brightness, in the simplic- 
ity of their auxiliary apparatus, and in the nature and quantity of 
their radiation. As these differences grew from the obvious limita- 
tions of the earlier lamps for studio use, they should be of immediate 
interest. 

Instead of the gravity-controlled pool of mercury limiting the 
position of the familiar Cooper Hewitt lamp tube, the new lamps 
have oxide-coated cathodes, the characteristics of which have been 
described elsewhere. 1 It will suffice here to say that a fused coating 
of alkali earth oxide, notably barium oxide, on nickel or tungsten, 
at a temperature of about 900 C. produces a low- voltage cathode of 
a general type adaptable to use either in high-vacuum electron de- 
vices, such as radio tubes, or in gaseous conduction devices, such as 
mercury, sodium, or neon lamps, the functioning of the cathodes be- 
ing largely independent of the nature of the surrounding gas. The 
one requirement for the proper operation of this type of cathode is 
the definite temperature already specified. Oxide-coated cathodes 
may be divided into two classes, depending upon whether their 
heating is solely that produced by the arc current and the voltage 
drop at the electrodes, or whether additional heating energy is sup- 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 
** Engineering Department, General Electric Vapor Lamp Co., Hoboken, N. J. 
110 



MERCURY AND SODIUM ARC LAMPS 111 

plied to the cathode from an auxiliary device. The arc-heated type 
of cathode happens to be used in the high-intensity mercury arc, 
whereas the separately heated type is used in current models of the 
sodium arc. The use of either type of oxide-coated cathode at once 
removes restrictions as to operating position, and so permits the 
same flexibility of use possible with incandescent lamps. 

The light-source length, or distance between the electrodes, of 
the new lamps is Ve to 1 / 8 that of standard Cooper Hewitt lamps, a 
fact that profoundly influences both their operating characteristics 
and their practical utilization. Electrically, the result is instantane- 
ous starting when cold, without any such electrical or mechanical 
starting means as ha\e usually characterized low- volt age arcs. The 
compactness results also in such a relatively high normal operating 
temperature and vapor pressure that the mercury arc does not re- 
start until it has cooled for an interval, which can be shortened by 
using one of the arc-starting devices long in use with gaseous con- 
duction lamps. 

Although both the sodium and mercury arcs can be operated on 
direct current, the fact that the more efficient types are respectively 
lower and higher in operating voltage than the conventional direct- 
current sources has led to a standardization at present on a-c. 
models. It is outside the scope of this paper to detail methods of 
arc operation other than to point out the additional practical ad- 
vantages of a-c. operation inherent in the voltage adjustment pos- 
sible with transformers, the efficiency of reactive rather than re- 
sistive ballasting, and the combination of those two factors with 
safety in the insulating line transformer with high-leakage reactance. 

In this connection, it should be noted that all arcs are inherently 
constant-current devices, and that arc-ballasting auxiliary devices 
are necessary only because constant-voltage rather than constant- 
current electrical distribution happens to be the practice. Series 
operation on constant-current lines is common in highway lighting 
practice and is being used with these new gaseous conduction arcs. 

When operated on a more or less constant-voltage line with re- 
active ballasting the lamps have a power-factor of about 0.65, which 
may be corrected by condensers, in the usual manner, to over 0.90, 
this residual apparent power-factor being due to a distortion rather 
than a time displacement of the current wave. 

The stability and regulation characteristics of gaseous conduction 
lamps can be made anything within reason, as they are fixed arbi- 



112 



L. J. BUTTOLPH 



[J. S. M. P. E. 



trarily by the amount of ballasting used. Stability; persistence of 
the arc upon sudden decreases in line voltage, and regulation ; changes 
in arc current or light, with changes in supply voltage; can be im- 
proved by increasing the ratio of the ballast voltage to the arc volt- 
age drop that is, by absorbing relatively more of the available 
voltage and energy in the ballast unit, with the obvious result of de- 
creasing the over-all efficiency. A similar design compromise is 
involved in the lumen-per-watt rating of a given gaseous conduction 
lamp. The efficiency may be increased by taking a shorter operating 
life, or a longer life may be had by operating at a lower efficiency. 
The complicated relation between arc-lamp stability and regula- 





FIG. 1. Sodium and mercury arcs 
without vacuum jackets or auxiliary 
equipment. Left: 10,000-lumen sodium 
arc. Center: 14,000-lumen high-in- 
tensity mercury arc. Right: 6000- 
lumen sodium arc. 

tion, over-all efficiency, and useful life has been discussed elsewhere. 2 ' 3 
It is to some extent true of incandescent lamps, but much more so 
of gaseous conduction lamps, that statements of either efficiency or 
life should always be in terms of the other factor and in terms of a 
stated or standardized method of operation. 

Although many other types and sizes are in development, only 
two sodium arcs have been standardized for manufacture and sale 
in this country (Fig. 1). A 10,000-lumen lamp rated at 200 watts 



Feb., 1935] MERCURY AND SODIUM ARC LAMPS 



113 




FIG. 2. Graphic indication of the energy required to produce comparable 
luminosities by light of various relative energy distributions. 



114 L. J. BUTTOLPH [J. S. M. P. E. 

in the arc and using about 25 watts in the auxiliary has an over-all 
efficiency rating of 45 lumens per watt when reactive ballasted for 
30 per cent of the total voltage and rated for a useful life of 1350 
hours. The light-source itself is 9 inches long and has an average 
brightness of 37 candles per square inch. 

A 6000-lumen lamp rated at 150 watts in the arc and using about 
15 watts in the auxiliary has an over-all efficiency rating of 36 lumens 
per watt when reactive ballasted for 30 per cent of the total voltage 
and rated for a useful life of 1350 hours. The light-source itself is 7 
inches long and has an average brightness of 34 candles per square inch. 

Both these sodium lamps must be enclosed in vacuum jackets to 
insure the temperature and sodium vapor pressure necessary for ef- 
ficient operation. For the same reason an initial interval of about 
20 minutes is required to attain an efficient operating temperature. 

Although a smaller percentage of the total energy put into sodium 
lamps emerges as visible radiation, the pair of lines are fortunately so 
placed relatively to the visibility curve, Fig. 2, that the luminosity 
is more than three times that of incandescent light, and hence their 
efficiency is 2 to 2.5 times that of a 20-lumen-per-watt incandescent 
lamp. 

This efficiency makes the lamp attractive for applications in which 
color values are of little importance, such as highway, flood, beacon, 
and spectacular lighting. 

Instead of distorting colors, sodium light renders only a limited 
range of yellow, and that with so high a luminosity as greatly to re- 
duce the apparent saturation. It has been suggested that this effect 
can be used to show visually the photographic black-and-white 
rendition of colored objects just as viewing filters are sometimes used. 
There is promise that the sodium lamp may be used for laboratory 
illumination in photographic processes using materials sensitive 
only to the blue and ultra-violet, but there seems to be little use for 
them as photographic studio light-sources with any of the photo- 
graphic emulsions now in general use. 

Although several sizes and types of high-intensity mercury and 
amalgam lamps are in actual development, only one, a high-pressure 
type of mercury arc, has been standardized for manufacture and sale. 
It is described as high-pressure because its normal mercury vapor 
pressure is about 1 atmosphere, or 1000 times that of the Cooper 
Hewitt lamp; and as high-intensity because the compactness and 
high vapor-pressure results in an increase in the intensity of the 577- 



Feb., 1935] 



MERCURY AND SODIUM ARC LAMPS 



115 



579 m/x lines relatively to the other lines, doubling the over-all arc 
efficiency, and a brightness 13 times that of the familiar Cooper 
Hewitt tube. 




FIG. 3. Graphic comparison of total energies, the relative energy distributions, 
and the luminosities characteristic of a-c. Cooper Hewitt low-pressure and 
General Electric high-intensity mercury arcs. 

This change in relative energy distribution with operating condition 
s shown graphically in Fig. 2 on the basis of equal luminosities, and 



116 



L. J. BUTTOLPH 



[J. S. M. P. E. 



in Fig. 3 on the basis of equal over-all wattages. In both figures, 

the total areas of the lines or 
blocks represent line energies, 
and the vertically ruled por- 
tions represent luminosities. 

The increased intensity of 
the yellow lines is paralleled 
by a slightly greater increase 
in the 366 m/* group, as 
shown in Figs. 2 and 3, and 
a slightly smaller increase in 
the 405 and 435 m/z violet 
lines, so that the relation be- 
tween foot-candles and photo- 
graphic effects is about the 
same as for the Cooper 
Hewitt lamp. 

In fact, the radiation char- 
acteristics of the high-inten- 
sity mercury arc are practi- 
cally the same as those of 
certain older high-pressure 
quartz mercury arcs, 4 except 
that the glass enclosures in- 
tercept the ultra-violet or 
short wavelengths. The 
high-pressure type of arc is 
also similar to the quartz 
mercury arc in its electrical 
operating characteristics, ex- 
cept that it is made more 
thermally stable by an in- 
sulating enclosure in an in- 
tegral outer bulb containing 
nitrogen at a pressure of 1 /z 
atmosphere, and by using so 
limited a quantity of mer- 

, cury as to assure complete 
FIG. 4. Comparison of various types of 

reflectors, used with a 425-watt high-inten- vaporization at the normal 
sity mercury arc. operating temperature. 




Feb., 1935] 



MERCURY AND SODIUM ARC LAMPS 



117 



The 14,000-lumen high-intensity mercury arc now available may 
be operated on alternating current at 220 volts with a series reactor, 
or at any voltage with a reactive transformer. The power-factor of 
65 per cent resulting from this manner of operation may be corrected 
to 92 per cent by using a condenser. The arc only is rated at 400 
watts, and the reactor and transformer may take an additional 25 to 
50 watts depending upon the design. The arc alone is rated at 35 
lumens per arc watt, when ballasted for 30 per cent of the total volt- 
age and rated for a useful life of 1500 hours. 

The light-source tube is 6 inches long and ! 3 /s inches in diameter. 
It, in turn, is enclosed in a cylindrical tipless bulb 2 inches in di- 
ameter, which is, with its mogul screw base, 13 inches in over-all 
length. This compactness results in an average brightness for the 

Low-press. Hg 
High-int. Hg 
Cd-Zn amalgam 
Mercury-inc. 
Mercury-neon 

Millimicrons 
FIG. 5. Spectrograms of mercury light modified or augmented as indicated. 




light-source itself of 175 candles per square inch. The unit is suf- 
ficiently compact for use with standard reflecting and diffusing equip- 
ment, and distribution curves possible with specially designed conical 
or cylindrical parabolic specular reflectors are shown in Fig. 4. 

The high-intensity mercury arc operates at so high a temperature 
that additional spectral lines can be attained by introducing small 
amounts of cadmium and zinc. The relative energy distribution 
of the red and blue lines thus added has been such as to produce a 
light slightly bluer in subjective color than the unmodified high- 
intensity arc, but still containing enough red to give slightly im- 
proved color discrimination. This change in color quality is ap- 
parently attained only at a sacrifice of efficiency, 5 as has usually been 
the case in connection with amalgam lamps. Although spectrograms 



118 L. J. BUTTOLPH [J. S. M. P. E. 

make only a qualitative comparison of spectra possible, those in 
Fig. 5 are a fair representation of the spectra obtainable from mercury 
and amalgam lamps. In this connection it should be noted that al- 
though it is not sufficient to show on the spectrograms of Fig. 5, 
the high-pressure high-intensity mercury arc contains enough red to 
make it noticeably better than the Cooper Hewitt low-pressure 
mercury arc in the discrimination of colors at the red end of the 
spectrum. 

Both the sodium and the high-intensity mercury arc as now avail- 
able operate as full-wave a-c. arcs having the flickering light char- 
acteristics of all full-wave a-c. arcs. The resulting stroboscopic 
effects in some applications may limit the use of the lamps, but there 
are still a large number of photographic processes in which it is no 
handicap. It is probable that high-intensity mercury arcs for d-c. 
operation will be available in case there is a definite need for their 
perfectly steady light. 

The arguments for the light quality of incandescent lamps and 
carbon arcs are generally more applicable to color photography than 
to the ordinary black-and-white rendition of colored objects; and 
there is plenty of evidence that for the economical production of 
high-quality black-and-white pictures, the mercury spectrum is still 
of great value. The bulk of the light-source and the bulk and weight 
of the auxiliary equipment have probably been more important than 
any other factors in the recent decrease in the use of Cooper Hewitt 
lamps in studio lighting practice. It is believed that the high- 
intensity mercury arc with its increased efficiency, decreased bulk, 
and the possibility of light direction and control by compact re- 
flectors, may find worth-while applications in modern studio light- 
ing practice. 

REFERENCES 

1 BUTTOLPH, L. J.: "A Review of Gaseous Discharge Lamps," Trans. Ilium. 
Eng. Soc., 28 (April, 1933), No. 2, p. 153. 

2 BUTTOLPH, L. J.: "The Electrical Characteristics of Commercial Mercury 
Arcs," Rev. Sci. Instr., 1 (Sept., 1930), No. 9, p. 487. 

3 BUTTOLPH, L. J.: "The Development of Gaseous Conduction Lamps," 
Trans. Electrochem. Soc., 65 (1934), p. 205. 

4 JOHNSON, L. B., AND BURNS, L.: "Line Intensity and Energy Distribution 
in High- and Low-Pressure Mercury Arcs," /. Opt. Soc. Amer., 23 (Feb., 1933), 
No. 2, p. 55.* 

8 WINCH, G. T., AND PALMER, E. H. : "A Method of Estimating the Proportion 



Feb., 1935] MERCURY AND SODIUM ARC LAMPS 119 

of Red Light Emitted by a Source, with Particular Reference to Gas Discharge 
Lamps," Illuminating Engineer (London), 27 (1934), No. 4, p. 123. 

DISCUSSION 

PRESIDENT GOLDSMITH: What is the approximate life of the lamps, in normal 
operation? 

MR. BUTTOLPH: The sodium lamps are rated at 1350 and the mercury lamps 
at 1500 hours. As stated in the paper, the rating involves a compromise with 
the efficiency at which the lamp is operated, and so is a tentative matter. 

MEMBER: Where are the street lighting installations located? 

MR. BUTTOLPH: One at Newton, Mass.; one at Revere Beach; one in New 
York City, at Jerome Avenue near Van Cortland Park; one in Pittsburgh, on 
Allegheny Boulevard. The largest I have heard of are being considered out 
on the West Coast. 

MR. CRABTREE: How long does it take to reach maximum intensity? 

MR. BUTTOLPH: A perfectly cold sodium lamp requires about twenty minutes. 

MR. CRABTREE: What is the shape of the intensity curve with time? 

MR. BUTTOLPH: Very low for the first five minutes; then it rises rapidly. 
This is a laboratory model, and the plate is a temporary support, you might say, 
for the vacuum jacket, or Dewar flask. You are seeing the sodium arc through 
three layers of glass. There is a bulb inside, and outside of it is a large Dewar 
flask similar to the unsilvered filler of a large-mouthed vacuum bottle. In the 
case of the mercury arc, the vacuum enclosure or the gas-filled enclosure is a 
part of the lamp structure. 

PRESIDENT GOLDSMITH: Would the lamps be in any way applicable, for 
example, to the modern effects in lighting the fronts of theaters? 

MR. BUTTOLPH: Yes, any place where the light may be used directly by re- 
flectors, where you want a highly efficient production of spectacular color. 

MR. CRABTREE: I can see a number of sun-spots in the bulb. What causes 
them? 

MR. BUTTOLPH: If your eyes become sufficiently fatigued you can see certain 
clouded areas of metallic sodium condensed on the inner bulb. We never vapor- 
ize all of it; sodium is always distilling around from one place to another hi the 
bulb. 

MR. CRABTREE: How does the mercury lamp start? 

MR. BUTTOLPH: Simply by applying the supply voltage; but it operates at 
such a very high pressure that it must cool for about ten minutes before it will 
restart. By adding a high-tension starting device, such as is used with the 
Cooper Hewitt lamps, it can be started much more quickly. For most applica- 
tions the delay in starting is not serious. As a matter of fact, the street lighting 
units will involve combinations of this lamp with an incandescent, and the incan- 
descent acts as a stand-by light source. 

MR. POPOVICI: What is the percentage of the near-ultra-violet to the visible 
rays, from the photographic standpoint? 

MR. BUTTOLPH: The actual energy is a little more than half that represented 
by the yellow and green light as shown graphically in Figs. 2 and 3. 

MR. PALMER : In the early days of the motion picture industry the low-intensity 



120 L. J. BUTTOLPH 

mercury vapor lamp was used extensively for studio lighting and a number of 
studios relied on it exclusively for illumination. The Astoria studio of Paramount 
Publix (opened in 1918) was equipped with 1000 tubes distributed in 120 banks. 
Despite the great bulk of the equipment and its weight, the lamps were much 
used because of their high efficiency and the desirable actinicity of the light. The 
introduction of improved arc and incandescent units and the need for directional 
light-sources gradually caused the mercury lamps to be displaced. 

The new high-intensity mercury lamps are more efficient than the old units 
and are more compact, and they undoubtedly have applications in the industry 
particularly for special problems. A recent application of these lamps was on a 
titling machine. The machine was equipped with a motor-driven camera running 
at 90 feet a minute, and the light-source consisted of two 35-ampere d-c. arcs 
placed about five feet away from the title board, one on each side. For these 
two arcs, taking a total of 7000 watts, four high-intensity a-c. mercury lamps 
having a total wattage of 1600 were substituted. It was found that the illumi- 
nation from the mercury lamps was sufficient to produce the same photographic 
results as before, with a considerable saving in current and maintenance. The 
increase in efficiency was, of course, not entirely due to the difference in the two 
types of light sources; there was a gain due to the fact that the mercury lamps 
could be placed closer to the title board, placing more of the light where it is 
needed. Due to the fact that the current alternated 120 times a second, and 
the camera made 24 exposures a second, there was no appreciable flicker in the 
projected picture. The new lamps should be useful in cartoon work also, as the 
lighting set-up is very similar to that required for titles. 

The mercury lamps, due to their high efficiency, produce very little heat as 
compared with other sources, and they are particularly suitable for under-water 
work or for work in enclosed spaces where good ventilation is not possible. 

MR. CRABTREE: Is there any difference in behavior between the a-c. and d-c. 
lamps? ' 

MR. PALMER: The flicker can be seen in the a-c. lamp, of course. The 
d-c. lamp is perfectly steady, and no stroboscopic effect is noticed when you move 
your hand in front of it. The light has the same characteristics except that the 
a-c. arc flickers and the d-c. arc doesn't. 

MR. CRABTREE: Is one of the lamps more apt to go out than the other? 

MR. PALMER: No; they remain lighted without any trouble. 

PRESIDENT GOLDSMITH: Does the 20-minute period during which the light 
intensity rises mean a corresponding change in exposure? 

MR. PALMER: Before the cameraman starts to load up the magazine he turns 
on the light, and it warms up while he is getting ready to work. The current is 
so small that it would cost only a few cents to leave it on all night. 

MR. HIBBEN: Better actinic effects occur when there is proportionately a great 
deal more energy hi the short wavelength region that affects the ordinary emul- 
sion. Over-all efficiencies may improve with larger sizes. It seems possible, 
and perhaps desirable, to concentrate tremendous amounts of light at certain 
spots. When we began to develop filament sources we went to 5-kw., 10-kw., 
and even 50-kw. lights in order to get high concentration for shadows and for 
speed. If there is an industrial need for anything like that with mercury sources, 
it seems possible, and very feasible but requires time for development. 



PIEZOELECTRIC LOUD SPEAKERS* 



A. L. WILLIAMS** 

Summary. Recent progress in piezoelectric loud speakers for theater use, includ- 
ing a new unit for upper registers, is described. Frequency-response characteristics 
of this unit alone and in combination are discussed and demonstrated. 

The character of the load upon the output circuit, and its effect upon the efficiency 
and the production of satisfactory response in the low and high registers, are discussed. 

The loud speakers described and demonstrated in this paper all 
operate on the piezoelectric principle, using bimorph elements of 
Rochelle salt as the means of converting electrical into acoustical 
energy. The construction of these crystalline units has been de- 
scribed in earlier papers 1 which have shown that the bimorph or two- 
layer principle is utilized to attain a mechanical magnification of the 
movements of the crystal under an applied voltage, and to minimize 
the undesirable effects of saturation, hysteresis, and temperature. 
The bimorph principle has been further developed so that multiple- 
layer elements are now used. 

Fig. 1 illustrates diagrammatically the construction of a typical 
general-purpose loud speaker operating on the piezoelectric principle. 
The motor consists of a double bimorph crystal element 2 x /2 inches 
square by x /4 inch thick, built up of four layers of Rochelle salt crys- 
tal slabs with electrodes attached to their surfaces. In the particular 
design illustrated, there is no electrode between the two inner slabs; 
therefore, as there is twice the thickness of material between the two 
inner electrodes compared with the outer layer, the electrical stress 
will be greater on the outside layers and will be commensurable with 
the mechanical motion. This mechanical and electrical stress equali- 
zation allows considerably more power to be handled than is possible 
with a two-ply unit. 

The crystal unit is supported inside a water-tight steel case by 
being held at three corners between pads. The fourth corner is 
fitted with a metal cap, an extension of which is brought out through 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 
** Brush Development Co., Cleveland, Ohio. 

121 



122 



A. L. WILLIAMS 



[J. S. M. P. E 



a water-tight seal for driving the cone through a ratio-arm, the pur- 
pose of which is to match the mechanical impedance of the cone to 
that of the crystal unit. The capacity of one of these speakers is 
about 0.01 /if, and they operate well from a source of 8000 to 12,000 
ohms. In fact, the response of the crystal is much better than that 
of the cone. Until that was realized there was a tendency to try to 
accommodate too wide a range of frequency on a single cone, and the 
advantage of using several units, each designed most efficiently for a 
particular portion of the audible spectrum, is now well recognized. 
This fact led to the design of the high-frequency electrophone, de- 
scribed by Ballantine, 2 which utilized a diaphragm composed of four 



RATIO 



CLAMPING 
PADS 




FIG. 1. Assembly of type R95 Piezoelectric reproducer. 



small bimorph units loaded by an exponential horn and designed to 
respond well up to 16,000 cycles. 

It is possible also to construct a highly efficient reproducer for the 
higher frequencies without a horn. Such a unit is now available, and 
consists of a crystal unit driving a cone. The most important part 
of the cone is solid and, in the present models, made of wood. The 
object of such construction is to provide equal distances for the sound 
waves to travel outwardly before meeting the air; and to overcome 
the very rough response due to interference that is noticeable with 
single paper cones. A cross-section of such a speaker is shown in 
Fig. 2, and a theoretical response curve of a properly matched type 
T-51 speaker and moving coil for theater combination in Fig. 3. The 



Feb., 1935] 



PIEZOELECTRIC LOUD SPEAKERS 



123 



speakers are small, light, and relatively inexpensive to construct, and 
are shallow enough to be mounted at the center of a loud speaker de- 
signed for low-frequency response. In fact, due to the efficiency of 
response of the crystalline elements in the upper register, the type 
T-51 speakers are produced at a cost sufficiently low that several 
units pointing in different directions may be very 
satisfactorily utilized in the same assembly to fur- 
nish a very uniformly distributed acoustic response 
at high frequencies. 

They are designed to be used directly across the 
output circuit of an electron tube in the case of 
radio receivers, and with a transformer in theater 
installations where a low-impedance or 500-ohm line 
connects to the speakers. Due to the high effici- 
ency of the reproducer, it will generally be found 
necessary to reduce the voltage so that the acous- 
tical output of the high-frequency unit is the same 
as that of the low-frequency unit. If a volume con- 
trol is inserted between the high-frequency and low- 
frequency speakers, it may be used as a very effec- 
tive tone control. By such a method the response 
of the elements actually reproducing the upper fre- 
quencies is increased or decreased so as to render 
the tones at their proper intensities. The high efficiency of the units 
results in a reserve of power sufficiently great actually to emphasize 
the high-frequency response, when required, without overloading the 
speaker or reducing the output of the other speakers in the circuit. 




FIG. 2. T-51 
high - frequency 
reproducer. 







100 1000 10000 

FIG. 3. Theoretical response curve of combined dynamic and type T-51 
theater speaker. 

As the speaker has a negative reactance, a filter is not required when 
it is used in combination with an inductive reproducer of the usual 
moving-coil type. In fact, it not only supplies the upper register, thus 
permitting the magnetic speaker to be designed for maximum effici- 



124 



A. L. WILLIAMS 



[J. S. M. p. E. 



ency over a. limited range, but also by its tendency to correct the 
power-factor of the dynamic speaker provides more efficient loading 
of the tube and circuit than would otherwise be obtained. The lead- 
ing-current characteristic of the piezoelectric speaker makes it very 
desirable as an output load, especially if it is necessary to use a step- 
down transformer to reduce the amplifier output to some line level. 
Carrying the idea of multiple speakers still further, a special theater 
speaker has been developed and is now available, built up from eight 



9 




R95-7 






Volume control for T SI S 

Connect I to /<?/7</2/ 2 L int. match iny front. 

Connect J to J and 4 to <f- 
FIG. 4. Circuit diagram of type 8RT theater reproducer. 






30 100 1000 

FIG. 5. Approximate distribution and total output of SRT reproducer. 



different speaker units designed to reproduce efficiently the whole 
acoustical range required for the new high-fidelity films. They are 
assembled in a single mounting and connected as shown in Fig. 4. 
The low- and middle-range speakers are connected in parallel, and 
are fed through a choke-coil across which are connected the high- 
frequency units. By properly designing the individual units and ad- 
justing the value of the series inductance, it is possible to design a 
speaker with a very flat response over the range required, as illus- 



: Feb., 1935] PIEZOELECTRIC LOUD SPEAKERS 125 

trated in Fig. 5. This speaker presents an impedance load on the 
amplifier which is virtually constant over this range. No field ex- 
citation is required, and due to the fact that the load is mostly ca- 
pacitive, a high-impedance line may be used without fear of losing the 
higher frequencies. The speakers are designed to operate across the 
plates of standard amplifier tubes. Where low-impedance lines al- 
ready exist between the projection room and the stage, a step-up 
transformer may be used to match the impedance of the speakers to 

I the line impedance. Two to four of these speakers may be used in 

the average size theater. 

REFERENCES 

1 SAWYER, C. B.: "The Use of Rochelle Salt Crystals for Electrical Reprodu- 
cers and Microphones," Proc. I. R. E., 19 (Nov., 1931), No. 11, p. 2020. 

BALLANTINE, S.: "A Piezoelectric Loud Speaker for the Higher Audio Fre- 
quencies," Proc. I. R. E., 21 (Oct., 1933), No. 10, p. 1406. 

DISCUSSION 

MR. MITCHELL: I have been informed that in some of these piezoelectric 
crystals there is a fatigue characteristic after they have been in operation for 
some time. Is that so? 

MR. WILLIAMS: We have had no sign of fatigue at all. We operated some 
speakers, in which a large piece of crystal was used, having a wooden arm at- 
tached to them to which a cone was fastened, for more than two years on 220 
volts, 60 cycles, in a little wooden shed where the humidity and temperature 
varied whiter and summer, and there was absolutely no sign of deterioration of 
the crystals. Installations of the earlier type of speaker have been operating in 
Severance Hall, the home of the Cleveland Orchestra, for five years, without a 
service call, either for microphone or speakers. The crystal must be enclosed 
however, otherwise in very hot, dry atmospheres there will be a tendency toward 
dehydration. 

MR. HICKMAN: What is the amplitude of the speaker cone? 

MR. WILLIAMS: The maximum amplitude so far attained is about an eighth 
of an inch. As used in the theater speaker, in which a number of units are used 
in a single baffle, each speaker loads the other, and the amplitude is reduced very 
greatly, thus enabling the speakers to handle considerably more power. The 
groups of eight will handle 20 watts safely, but 5 watts are recommended as a 
working maximum. The individual loading between the cones restricts the 
motion and, incidentally, due to the back emf., raises the impedance quite con- 
siderably. 



REFLECTING SURFACES OF ALUMINUM* 



J. D. EDWARDS** 

Summary. Aluminum is inherently a good reflector of radiation in the ultra- 
violet, visible, and infra-red portions of the spectrum. Means of developing and main- 
taining this high reflectivity in commercial reflectors are described. The new Alzak 
process of electrolytic brightening and oxide coating makes available aluminum re- 
flectors, both specular and diffuse, having reflection factors up to about 85 per cent. 
The Alzak reflectors have a hard, abrasion resisting coating of clear transparent 
aluminum oxide on the surface which makes them weather-resistant and easily 
cleaned. 

Aluminum, the metal, is intrinsically bright. To make a good 
practical reflector from aluminum requires the solution of two prob- 
lems : first, to develop suitable methods for bringing out or develop- 
ing the inherent brightness of the metal ; second, maintain the bright- 
ness of the reflecting surface under service conditions. 

For brightening a metal surface, the obvious and time-honored ex- 
pedient is to polish it. Where a specular surface is required this 
operation may be necessary, but it has certain limitations. With 
commercial aluminum sheet, about the best reflectivity that can be 
attained by polishing is of the order of 65 to 75 per cent. It is true 
that higher reflectivities, even up to 89 per cent, have been achieved 
by polishing some of the hardest aluminum alloys by a special technic, 
but such a reflecting surface can have only very limited application. 

Where a diffusing surface is satisfactory, chemical etching methods 
are very effective in bringing out the high reflecting power of alumi- 
num. 1 One such etching medium is an aqueous solution containing 
about 5 per cent sodium hydroxide and 4 per cent sodium fluoride. 
The surface is etched and brightened by immersion in this solution 
at a temperature of about 90 C. A final dip is given in a cold solu- 
tion of nitric acid containing equal parts of concentrated nitric acid 
and water. Using this method of etching, aluminum diffuse reflect- 
ing surfaces are being commercially produced, which have a reflection 
factor for visible light ranging from 82 to 87 per cent, and for 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 
** Aluminum Research Laboratories, New Kensington, Pa. 
126 



REFLECTING SURFACES OF ALUMINUM 



127 



ultra-violet radiation at a wavelength of 296.7 m/* of 81 to 82 per cent. 

Fig. 1 shows a curve of the reflectivity of aluminum for wave- 
lengths ranging from 0.2 to about 12 /*.* This curve represents values 
that are practically attainable, and is interesting for a number of 
reasons. Aluminum, it will be seen, is an excellent reflector of ultra- 
violet radiation; it is unexcelled in this respect by any other metal. 
This characteristic is of special significance in the photographic art, 
because of the sensitivity of photographic film to short-wave radia- 
tion. The reflectivity of aluminum is nearly 90 per cent in the 
visible range, and attains values as great as 97p er cent in the infra-red 
range. 

Having produced a surface of high reflectivity by the etching pro- 



100 
90 

Of) 


Ultra-violet- 


Visible 
Alight -~|- 


Infra-red 






^ 


** 


^ 


=0= 


= 


^^ 


- 


= 
















































70 

l_ CA 






/ 












































/ 


























o 





A 


fay lor and 
Edwards 
Coblentz. 
Hulburt. 


Edwards. 


1" 

540 
'o30 

4> 
































































i 








































































^^ U 

cc 10 

























































































J .2 3 4 5 J6 .7 B .9 ID 


234 56789 10 II 1 



FIG. 1. 



Wave Length of Incident Light.-^i 

Reflectivity of aluminum for radiations of various wavelengths. 



cedure just outlined, the problem remains of maintaining the surface 
against depreciation under severe service conditions. Indoors, 
where frequent cleaning is not required, etched reflectors of this type 
have proved highly satisfactory. Where they collect dust and dirt 
rapidly, however, or where they are subject to weathering, such a 
surface is not as readily cleanable as might be desired, and may show 
considerable surface attack when exposed continuously. The sur- 
face can be protected with a clear, colorless lacquer, but only at a 
sacrifice of some 10 per cent in reflection. Outdoors, of course, the 
lacquer protection is relatively short-lived, and may become yellow 
and discolored if not carefully selected. 
Another type of protection that is available is oxide coating of the 



128 J. D. EDWARDS [J. S. M. p. E. 

aluminum surface. Electrolytic or anodic oxidation of aluminum, 
as typified by the Alumilite process, is a means of providing aluminum 
with a hard and abrasion-resisting coating of aluminum oxide. This 
coating is likewise highly impervious to moisture and is protective 
against weathering. An oxide coating, however, that is thick enough 
to confer all these properties in the desired degree decreases the re- 
flectivity of the surface from about 10 to 20 per cent, and leaves the 
metal with a rather opaque and milky surface finish which naturally 
detracts from its value as a reflector. 

A discovery of R. B. Mason of Aluminum Research Laboratories 
has solved in a highly satisfactory way the problem of producing 
bright aluminum surfaces and maintaining them under service con- 
ditions. The method is that of electrolytically brightening alumi- 
num so as to bring out its maximum reflectivity. In this process, the 
aluminum reflector, formed to final size and shape, is made the anode 
in an electrolyte of novel composition, acting in which capacity the 
reflector is brightened by the electrolytic removal of impurities, both 
metallic and non-metallic, from the surface of the metal. This proc- 
ess is unique in that the brightening is effected without any etching 
or roughening of the surface; thus, a highly polished reflector can be 
electrolytically brightened without appreciable loss of specularity. 
For example, a polished aluminum reflector having a reflection factor 
of 74 per cent, after being subjected to this electrolytic brightening 
treatment, had a reflection factor of 87 per cent. A very thin pro- 
tective oxide film is formed upon the surface, but it is too thin to with- 
stand many service conditions satisfactorily. Very fortunately, how- 
ever, the electrolytically brightened surface can be further oxidized 
by the Alumilite process to give it a substantial protective coating 
of oxide without any important loss in reflectivity. Specular re- 
flectors having reflectivities of 80 to 85 per cent are being commer- 
cially produced by this process. Application of the final oxide film 
appears to cause a somewhat greater loss of reflection for ultra-violet 
and infra-red radiation than for visible light. 

The trade-name Alzak has been applied to this process and to re- 
flectors made by it. The process in outline consists first in preparing 
the surface by polishing or etching, depending upon whether a specu- 
lar or diffusing surface is required. The surface is then electrolyti- 
cally brightened and oxidized, whereupon it becomes a bright reflector 
having a clear, transparent oxide coat upon its surface. As a final 
treatment, the oxide-coated reflector is "sealed" so as to make the 



Feb., 1935] REFLECTING SURFACES OF ALUMINUM 129 

oxide coat impervious and resistant to staining or marking by hand- 
ling or in service. For the best results, aluminum sheet of special 
type and composition is selected for reflectors to be finished by the 
Alzak process. The Alzak reflectors have a smooth, hard, glassy 
surface which does not collect dirt and which can be readily cleaned 
by washing with soap and water. If a more thorough cleaning is 
necessary, the use of a mild abrasive such as Bon- Ami is satisfactory 
and cleans without injury to the surface. 

During the past year there has been an unusual interest in alumi- 
num reflectors, both because of the development of the Alzak process 
and also because of the application of evaporated aluminum films to 
telescope reflectors. In this latter application, instead of preparing 
a highly reflecting surface upon aluminum, a very thin film of alumi- 
num is deposited upon the glass or metal surface to be used as the 
reflector, producing what is known as a "first-surface mirror." Alumi- 
num normally has a boiling point at atmospheric pressure of about 
1800 C. If, however, the pressure is reduced to a very low value, 
the boiling point is correspondingly reduced. For example, at a pres- 
sure of about 0.001 mm. of mercury the boiling point of aluminum is 
about 730 C, and at a pressure of 0.1 mm. is about 950 C. If, there- 
fore, an object to be coated is suspended in a highly evacuated cham- 
ber, and aluminum contained therein is electrically heated to its 
boiling point, the aluminum will evaporate and be deposited as a 
bright metallic film upon any surface in the path of the stream of 
aluminum vapor. Such films are apparently highly satisfactory for 
telescope reflectors because of their high reflectivity for the photo- 
sensitive ultra-violet radiation and also because of their non-tarnish- 
ing characteristics. Such films are not, however, able to withstand 
the handling or cleaning to which Alzak reflectors can be subjected 
nor are they weather-resistant. 

Aluminum paint is another medium by which bright reflecting 
surfaces may be produced. It has been extensively used for motion 
picture projection screens. The reflection factors of aluminum- 
painted surfaces will vary from 60 to 75 per cent, depending upon the 
vehicle and upon the grade of aluminum bronze power employed. 
The smoothness of surface and diffusiveness can be controlled to some 
extent by proper formulation of the paint. Aluminum bronze 
powder is also available in the form of a paste, which, with a suitable 
vehicle, gives a bright and uniformly diffusing surface. The luster 
of an aluminum-painted screen adds "life" to the projected picture. 



130 J. D. EDWARDS [J. S. M. P. E. 

An important application of the Alzak process is in the production 
of lighting reflectors. The Alzak reflector has the advantages of 
lightness and non-breakability in addition to its high reflectivity 
and non-tarnishing characteristics. Aluminum reflectors can also 
be formed to the required shape, with the high precision that is neces- 
sary in some forms of projectors. Tests are under way to develop 
the possibilities of this process in preparing optically accurate re- 
flecting surfaces, both small and large. It is too early, however, to 
make any prophecies as to what can be done in this very exacting 
field. 

Another potential application of the Alzak process is in the pro- 
duction of small projection screens for home use. The bright alumi- 
num surface made suitably diffusing presents an attractive back- 
ground for the picture, and its lightness and ready washability are 
additional desirable characteristics. Other applications of bright 
metal trim and surfaces naturally suggest themselves. It is apparent 
that the motion picture industry may find many uses for these new 
products. 

REFERENCES 

BAYLOR, A. H., AND EDWARDS, J. D.: "Ultraviolet and Light Reflecting Prop- 
erties of Aluminum, /. Opt. Soc. of Amer., 21 (Oct., 1931), No. 10, p. 677. 

2 EowARDS, J. D.: "Aluminum for Reflectors," Trans. Ilium. Eng. Soc., 29 
(May., 1934), No. 5, p. 351. 

TAYLOR, A. H.: "Reflection Factors of Various Materials for Visible and 
Ultraviolet Radiation," /. Opt. Soc. of Amer., 24 (July., 1934), No. 7, p. 192. 

DISCUSSION 

MR. SANDVIK: What methods are used for polishing aluminum surfaces? 

MR. EDWARDS: Aluminum reflectors are usually finished by buffing with 
"white diamond" or similar polishing compounds. 

PRESIDENT GOLDSMITH: Do these two samples differ in their resistance to 
finger-marking, for example? 

MR. EDWARDS: One is finished by buffing and the other is oxide-coated. The 
buffed sample will finger-mark readily. The other may show finger-marks, but 
they will wipe or wash off. You can not satisfactorily remove them from the first 
reflector except by rebuffing. 

PRESIDENT GOLDSMITH: The same distinction holds as well for the two 
etched reflectors? 

MR. EDWARDS: Yes. You can tell the difference at once by running your 
finger over the surfaces; the oxide-coated surface has a smooth, glassy feeling. 

MR. CRABTREE: What is the effect of gases present in the air, such as 
sulfur dioxide, sulfuric acid, hydrogen sulfide? 



Feb., 1935} REFLECTING SURFACES OF ALUMINUM 131 

MR. EDWARDS: The oxide coatings are not resistant to high concentrations of 
many chemical reagents. We have had reflectors with the Alzak surface exposed 
to the New Kensington industrial atmosphere for a period of over a year with 
no loss of reflection factor. When brought into the laboratory and washed, 
they exhibited their original reflection. They will, of course, collect dirt. 

MR. CRABTREE: How does the reflection factor of a lamp reflector fall 
off under practical conditions without cleaning? Is the loss of reflective power 
due entirely to dirt? 

MR. EDWARDS: In the case of the Alzak aluminum reflector it is due almost 
entirely to dirt. In one case I know of, measurements were made on a silver- 
plated reflector and an Alzak reflector; the silver-plated reflector had been plated 
and polished three or four days before. Measurements were made of the two, 
and they were then cleaned and re-measured. There was no increase or decrease 
in reflection of the Alzak reflector, but cleaning increased that of the silver-plated 
reflector by 5 or 6 per cent, showing the depreciation that had taken place in a 
matter of three or four days. Aluminum is not darkened by sulfides as is 
silver. 

MR. FRIED: In the production of motion picture screens, can you control the 
diffusion characteristics very accurately? 

MR. EDWARDS: Only approximately. 

MR. JOY : How high a temperature can the reflector withstand without becom- 
ing discolored? 

MR. EDWARDS: In some of our experimental work the temperature of the re- 
flector surface was about 320 F. over quite an extended period. There was no 
depreciation. 

MR. JOY: Have you done any work at, say, 200C.? 

MR. EDWARDS: Some experiments have been reported in which Alzak re- 
flectors were heated in an oven for 500 hours at 250 C. with only a small loss of 
reflectivity. 

MR. DAY: How large sheets can you handle now, particularly in the anodic 
treatment? 

MR. EDWARDS: We have only a small plant at New Kensington for the Alzak 
process, but we have coated sheets about three feet square. 

MR. DAY: Have you attempted 10 X 12 feet? 

MR. EDWARDS: Not yet; but it is possible if you have the tank equipment for 
it. 

MR. PALMER: Can you make as good a reflector by the vaporization process 
as you can by the anodic process? 

MR. EDWARDS: My knowledge of the vaporization process is second-hand, 
but I have a reflector made on glass that has a reflection factor of 90 per cent. I 
have seen what I thought were reliable measurements going even above that, say, 
92 per cent. You can make a better reflecting surface by that process than by the 
Alzak process. However, it will not withstand the service conditions that this 
Alzak reflector will, because of the thin, necessarily soft, film of aluminum. But 
it is very serviceable on the reflector of an astronomical telescope. This surface 
would be of particular value in stellar photography where you have the high re- 
flection factor of aluminum for ultra-violet light, giving an increased sensitivity 
in photographically recording the light from stars. 



132 J. D. EDWARDS 



MR. RAVEN: Have tests been made showing the reflection factor and dif- 
fusion characteristics of this finish on flat sheets? If so, how does it compare 
with the surfaces we now use for motion picture screens? Also, would it be 
difficult to keep a screen with this finish clean and efficient? 

MR. EDWARDS: Maintenance would be relatively easy. Any ordinary kind 
of dirt is readily removed by washing with soap and water, and the surface will 
dry free from water-marks. 

MR. RAVEN: In other words, ordinary washing would restore the surface to 
practically its original condition? 

MR. EDWARDS: There would be no difficulty whatever. 

MR. GAGE: I have had the good fortune to see this process of evaporating 
aluminum on glass. It is done at the Physical Laboratory of Cornell University 
by Robley C. Williams. If you start with a good, smooth glass surface, you can 
get a most excellent coat of aluminum. However, there is no advantage in 
coating the aluminum so thickly that you can not see the filament of an incandes- 
cent lamp through it. The coating can be made so permanent that you can scrub 
the aluminum coat with cheese cloth, for perhaps 500 times, just as hard as pos- 
sible, without apparently injuring the surface. This extra durability is secured 
by first evaporating a thin layer of chromium on hard Corning borosilicate glass 
followed immediately by a layer of aluminum evaporated from another filament. 
When removed from the vacuum the aluminum is hardened by washing in tap 
water, alcohol, or by condensed breath moisture. 

How much heat can the all-aluminum reflectors withstand? That is probably 
a matter that is exciting considerable interest at the present time among the 
projection engineers. As we know, the back-surface silvered glass reflector will 
not stand heat above a temperature at which the backing breaks down. We 
shellac the surface and gain a certain resistance to heat, which is small. If the 
back of the silver is copper-plated and we use a high-grade varnish, then we can 
use higher temperatures. 

What particularly interests the projection engineer is whether the specular 
aluminum surface, as treated, can stand a temperature comparable with that 
withstood by a silvered reflector backed with a good enamel. 

MR. EDWARDS: I can not say just what temperature can be withstood with- 
out depreciation. The aluminum oxide on the surface is not easily discolored by 
heat. When the metal and oxide have been stretched sufficiently by thermal 
expansion, hairline cracks appear. They do not injure the reflection efficiency 
of the surface appreciably, and are so fine that they can not be seen when viewed 
directly. The light must be almost at the angle of grazing incidence in order to 
see them. Our hopes, at least, are that reflectors of this type will stand reason- 
ably high operating temperatures, and do so very satisfactorily. 

MR. MITCHELL: Have you any comparative data on the reflection character- 
istics of chromium, for instance, and this new finish? 

MR. EDWARDS: The data I have seen from Nela Park indicate that chrome- 
plate has a reflection factor of the order of 60 to 65 per cent ; with the new finish 
the reflection may be 80 to 85 per cent. 



SIMPLE THEORY OF THE THREE-ELECTRODE 
VACUUM TUBE* 

H. A. PIDGEON** 



Summary. The physical principles upon which the operation of the three-ele- 
ment vacuum tube depends are presented in simple form and the terms usually 
applied to the tube, its operation as an amplifier, and a simple approximate method 
for computing the power output and percentage of distortion are explained. 

No new material is presented in the paper although some of it is presented from a 
somewhat different point of view from that usually found in the literature. An 
effort has been made to present in reasonably compact form the essential features of 
the subject most useful to engineers interested in vacuum-tube applications. 

The subjects discussed include: the portion of electron theory upon which the 
fundamental principles of vacuum-tube operation are based; space charge, the 
three-halves power law, temperature and voltage saturation; characteristics of the 
three-element tube; definition and physical significance of the terms plate resistance, 
transconductance, and amplification factor; dynamic characteristics, power output, 
and distortion; various means of coupling the vacuum tube to its associated circuits; 
and means for testing vacuum tubes for adequate thermionic emission. 

Vacuum tubes depend for their operation upon the flow of a stream 
of electrons through the evacuated space between the electrodes of 
the tube. To produce such a flow of electrons, which constitutes an 
electrical current, two things are necessary: First, there must be a 
continuous source of supply for the electrode producing the flow of 
current. The electrode producing this continuous supply of electrons 
is designated the cathode. It must be maintained at an elevated 
temperature necessary for the liberation of electrons from it. Second, 
to produce a continuous flow of the electrons, some force must be 
supplied to propel them. Since they are very small negative charges 
of electricity, they are attracted by a positively charged body and re- 
pelled by a negatively charged body. Consequently, if a second elec- 
trode within the vacuum tube be maintained at a positive potential 
with respect to the cathode, it attracts the negatively charged 
electrons to it at a rate depending upon the rate at which they are 

* Presented at the Spring, 1934, Meeting at Atlantic City, N. J. 
** Bell Telephone Laboratories, Inc., New York, N. Y. 

133 



134 H. A. PIDGEON [J. S. M. P. E. 

supplied by the cathode, or upon other factors which will be dis- 
cussed later. Such an electrode, maintained at a positive potential 
and acting as a collector of electrons, is designated the anode or plate. 

It is necessary that the electrodes be inclosed within an evacuated 
envelope, because the presence of air or other gases, even at low 
pressures, seriously interferes with the performance of the vacuum 
tube. Such a vacuum tube as we have described, consisting of an 
evacuated envelope containing two electrodes, a cathode acting as an 
emitter of electrons, and a positively charged anode or plate acting 
as a collector of electrons, is called a diode or two-electrode vacuum 
tube. Obviously, when the plate of such a tube is maintained at a 
negative potential with respect to the cathode, it becomes negatively 
charged, and consequently repels the electrons so that it ceases to 
collect them and no current flows. The electron current flow in 
such a tube is, then, unidirectional; that is, it flows only from the 
cathode to the anode, and only when the potential of the anode is 
positive with respect to the cathode. If an alternating voltage, 
or one that periodically changes from positive to negative values and 
vice versa, be applied between the electrodes of such a tube, current 
will flow in the tube only during the portion of the cycle when the 
plate is positive with respect to the cathode. Such a device is capable, 
then, of converting alternating current into unidirectional or direct 
current. This function is performed by the rectifier tubes employed 
in all modern radio receivers designed to operate on alternating- 
current supply. 

Furthermore, in such tubes resistance is offered to the flow of 
electrons, the nature of which will be discussed more fully later. 
Tt will suffice to state here that its origin is in the electrical field set 
up in the space between the cathode and anode by the negative 
electrons, and that its magnitude depends upon the size, shape, and 
spacing of the electrodes. If the rate of supply of electrons by the 
cathode is sufficiently large, this resistance limits the magnitude of 
the current that may flow at any given anode potential. The current 
increases with increasing positive plate potential and, consequently, 
in such a tube it can be varied by varying the potential of the plate. 

What is desired in most applications of vacuum tubes, however, 
particularly in amplifiers, is some means of controlling the flow of 
electrons to the anode in accordance with variations of an input 
signal. It is further desired to accomplish these variations in current 
to the anode with the expenditure of the least possible amount of 



Feb., 1935] 



THREE-ELECTRODE VACUUM TUBE 



135 



energy, so that it will be possible for very weak signals to perform 
this control function. 

Such a means of control is found by the insertion of a grid or mesh 
structure between the cathode and anode of the two-electrode tube 
described above. By maintaining the grid at some appropriate 
negative potential, it will repel electrons and will in part, but not 
wholly, neutralize the positive or attractive force exerted upon them 
by the positive plate. Consequently, an electron stream will still 
flow through the grid to the plate, 
but it will be smaller than it would 
be if the negative grid were not 
present. By making the grid less 
negative, its repelling effect will be 
reduced and a larger current will 
flow through it to the plate. In a 
similar manner, if the grid is made 
more negative, its effect will be 
increased and the current to the 
plate will be correspondingly re- 
duced. If the potential of the 
grid be made to vary about some 
mean value in accordance with 
some desired signal, the plate cur- 
rent will vary in a corresponding 
manner. Since the grid is assumed 
at all times to be at a negative 
potential with respect to the 
cathode, it can not collect elec- 
trons, and so, a very minute 
amount of energy will suffice to 
vary its potential in accordance 
with the input signal. 

Since such a vacuum tube has three electrodes, viz., cathode, 
anode, or plate, and grid, it is commonly referred to as a triode or 
three-electrode vacuum tube. The arrangement of the electrodes 
in a three-electrode tube is shown in Fig. 1. To maintain the elec- 
trodes of such a tube in an operating condition and at the proper 
potentials requires three sources of energy or potential as follows : 

To maintain the cathode at the high temperature necessary for the 
liberation of electrons from it requires energy to be supplied to it. 




GRID BATTERY 



PLATE BATTERY 



FIG. 1. Schematic diagram of a 
triode, showing battery connections. 



136 H. A. PlDGEON [J. S. M. P. E. 

If the cathode is in the form of a wire or ribbon filament heated 
by passing current through it, the source of electrical energy, if it be a 
battery, is referred to as the filament battery. The arrangement of 
such a battery is shown in Fig. 1. The voltage applied to the fila- 
ment terminals is designated by E a . Similarly, batteries supplying 
the positive plate potential and the negative grid potential, or grid 
bias, are referred to as the plate battery and grid battery, respectively. 
The arrangement of these batteries is shown also in Fig. 1. Having 
briefly described the elements of the triode and outlined its essential 
fundamental principles, we shall next proceed to describe these 
features in more detail and then show how the device operates as an 
amplifier. 

ELECTRONS 

All matter is made up of submicroscopic particles. These particles, 
which are the smallest into which matter can be subdivided 
and still retain the properties of the original substance, are called 
molecules. Molecules of different substances vary greatly in com- 
plexity, ranging from extreme simplicity in some substances to very 
great complexity in others. Ultimately, however, all molecules 
may be broken up into simpler constituents called atoms. Of these 
there are about ninety distinct kinds known, each representing 
one of the chemical elements from which all matter is constructed. 
Only a few elements, however, appear in the molecules of any one 
of even the most complex substances. An element, then, is a funda- 
mental substance composed of only one kind of atom. In some ele- 
ments, the molecules are composed of single atoms; in other 
elements, two or more like atoms are associated together to form 
the molecule. Some of the more common elements are hydrogen, 
oxygen, nitrogen, carbon, iron, nickel, copper, etc. 

Carrying the analysis further, atoms are well known to have com- 
plex structures. According to the most widely accepted modern 
physical picture of the atom, it corresponds roughly to a miniature 
solar system. Corresponding to the sun in our solar system is the 
nucleus of the atom which, in general, is a very small, compact 
structure composed of a combination of extremely minute particles 
called protons, neutrons, positrons, and electrons. The proton, whose 
mass may be taken as the unit of atomic weight has a positive 
charge equal in magnitude, but opposite in sign, to that of the elec- 
tron. Its mass is very large compared with that of the electron 
or of the positron. The neutron has very nearly the same mass as the 



Feb., 1935] THREE-ELECTRODE VACUUM TUBE 137 

proton, but is uncharged. The positron may be regarded as the 
ultimate unit of positive charge just as the electron is the ultimate 
unit of negative charge. The positron has the same magnitude of 
charge as the electron and very nearly (at least) the same mass. 
Practically all the mass of the atom is associated with the small, 
dense nucleus. Revolving about the nucleus in orbits at relatively 
large distances from it, are one or more electrons. 

The simplest of all atoms is that of hydrogen, whose nucleus con- 
sists of a single proton with a single electron revolving about it. 
The mass of the nucleus in this case is 1840 times that of the electron. 
The next atom in simplicity is that of helium, whose nucleus consists 
of four protons and two electrons bound together in a compact cen- 
tral core of great electrical stability. Revolving about this compact 
nucleus are two electrons. 

The atoms of the other elements become increasingly more com- 
plex by the successive addition of one electron to those revolving 
about the nuclei, and with the progressive addition of protons, 
neutrons, positrons, and electrons to the nuclei. In every case the 
normal atom has an exactly equal number of positive and negative 
elementary charges, so that the atom as a whole is neutral; that is, 
it behaves toward electrified bodies at some distance from it as though 

it had no charge at all. 

CONDUCTION 

From the standpoint of this discussion, the atoms of certain sub- 
stances, known as conductors, exhibit a very important property. 
This is particularly true of the metals copper, aluminum, silver, 
platinum, iron, nickel, etc. The outermost electrons in the atoms 
of these materials are so loosely attached to the atoms that they 
actually escape and wander from atom to atom. These wandering 
electrons are called free electrons, although the individual unattached 
electrons probably remain so for only very short intervals. How- 
ever, in the aggregate the free electrons per cubic centimeter at any 
instant amount to an extremely large number. They are in a state of 
continual rapid motion, or thermal agitation. The situation is analo- 
gous to that in a gas where it is known that the molecules, according 
to the kinetic theory, are in a state of rapid motion, with a random 
distribution of velocity. Now, if it were possible at a given instant 
to examine the individual molecules or electrons, it would be found 
that their velocities vary enormously (theoretically they vary from 
zero to infinite velocity), but that the average kinetic energy of the 



138 H. A. PlDGEON [J. S. M. P. E. 

molecules or electrons is constant for a given temperature, and that 
it varies with the temperature, being in fact directly proportional 
to the absolute temperature. 

If a conductor, let us say a copper wire, is placed in an electrical 
field that is, if the ends of the wire are maintained at different po- 
tentials by means of a battery there is a slow drift of the "atmos- 
phere" of free electrons along the wire toward the end at the higher 
potential. This slow average drift, which is superimposed upon the 
relatively very rapid random motion of the individual electrons due 
to thermal agitation, accounts for the usual conduction currents 
observed in conductors. 

THERMIONIC EMISSION 

Another important phenomenon that depends upon the free elec- 
trons in conductors is thermionic emission. This is the passage of an 
extremely small fraction of the free electrons through the walls of the 
conductor into the space outside. Since the free electrons have high 
average velocities, they would escape in large numbers even at ordi- 
nary room temperatures were it not for the fact that at the surface 
of metals there are very strong forces tending to pull the electrons 
back into the interior. The situation may be illustrated by the 
following rough analogy: Suppose that we have a rather deep box 
open at the top and partially filled with small rubber balls. If the 
balls are kept in a state of agitation, let us say by shaking the box, 
an occasional ball will acquire enough velocity, in the proper direc- 
tion, to escape from the box. That is, by a particularly favorable 
series of collisions with other balls, it acquires enough kinetic energy 
so that it may perform the definite amount of work required to lift 
itself over the walls and escape. If the balls are more and more 
vigorously agitated, correspondingly more of them will acquire 
enough energy to escape. 

Just as the height of the walls of the box represents a definite 
amount of work the rubber balls must do to escape, so, too, the sur- 
face forces in a conductor represent a definite amount of work an 
electron must perform to escape from the conductor. This work is 
known as the thermionic work-function. It is sometimes expressed in 
terms of the equivalent number of volts; that is, the potential differ- 
ence, in volts, through which an electron must fall to acquire a value 
of kinetic energy equal to the work-function. It is different for 
different substances. For example, in tungsten it is 4.52 volts. 



Feb., 1935] 



THREE-ELECTRODE VACUUM TUBE 



139 



In all known substances the work-function is so large that sub- 
stantially no electrons can escape at ordinary temperatures. How- 
ever, as the temperature is raised the average kinetic energy of the 
free electrons is increased proportionally until finally a temperature 
is reached at which an appreciable number of electrons begin to 
escape. In pure metals this temperature varies from 1600 to 1800 K. 
The rate of emission of electrons per unit area (sq. cm.) as a function 




200O 2200 2400 2600 

TEMPERATURE IN DEGREES KELVIN 



FIG. 2. Electron emission from one square centimeter 
of pure tungsten as a function of temperature. Curves 1, 
2, and 3 show the emission current in microamperes, milli- 
amperes, and amperes, respectively. 



of temperature follows a law given by what is known as Richardson's 
equation. It is* 



= AT 1 /* 



(1) 



* W. O. Richardson, S. Dushman, and others have derived a slightly different 
expression for electron emission. It is: 

Is = A T\ -*' e / kT 

in which A theoretically has the same value for all metals. Either equation may 
be taken as a satisfactory expression for electron emission. In fact, it is very 
difficult to obtain emission data with sufficient precision to distinguish between 
the two equations. 



140 



H. A. PlDGEON 



[J. S. M. P. E. 



in which I s is the emission current in amperes per square centimeter, 
is the Napierian logarithmic base, k is the so-called Boltzmann 
constant (1.37 X 10 ~ 23 joules per degree), A is a constant depending 
upon the emitting substance, <p is the work-function (in volts), e is 
the charge of the electron (1.59 X 10 ~ 19 coulombs), and T is the 
temperature (K.). 




WATTS PER CM 2 



25^070 



FIG. 3. Electron emission from one square centimeter of tung- 
sten and oxide-coated filament, as a function of the heating energy 
in watts per square centimeter. 



The emission from one square centimeter of pure tungsten is 
shown by the curves of Fig. 2. Curves 1, 2, and 3 give the current 
in microamperes, milliamperes, and amperes, respectively. They 
show how extremely rapidly the emission increases with tempera- 
ture. 



Feb., 1935] THREE-ELECTRODE VACUUM TUBE 141 

Since the energy required to heat a thermionic emitter or cathode 
is dissipated by radiation from the surface, excepting what is lost by 
conduction through the leads, it is convenient to express the emission 
as a function of the number of watts dissipated per unit area of the 
surface. Curve 1 of Fig. 3 shows the emission from one square centi- 
meter of tungsten plotted in this manner. The coordinates are so 
chosen that emission data following Richardson's equation give 
a straight line. 

Certain oxides, such as barium and strontium oxides or a combina- 
tion of the two, when applied to the surface of a metal, have the 
property of enormously increasing the electron emission at a given 




QQflQO 



FILAMENT 
TRANSFORMER 



FIG. 4. Schematic diagram of circuit connections to a triode employing 
a filament transformer. 



temperature. This is due to a very material decrease in the work- 
function. For example, the work-function is equivalent to about 4.52 
volts for pure tungsten, and to 1.0 volt or less for oxide-coated fila- 
ments. The emission from such filaments varies widely, but curve 2 
of Fig. 3 may be taken as a typical illustration of emission lines 
obtained for one square centimeter of coated filament. The great 
difference between it and tungsten is at once apparent. 

For example: Tungsten operating at 2400 K. emits 116 milli- 
amperes per square centimeter of surface and requires about 58 watts 
per square centimeter to heat it. The emission is, then, 2 milli- 
amperes per watt. On the other hand, oxide-coated filaments 
operate within the range 950 to 1125K., with a power dissipa- 
tion of 2.0 to 5.0 watts per square centimeter, and an emission of 75 
milliamperes or more per watt. 



142 



H. A. PlDGEON 



[J. S. M. P. E. 




CATHODE SLEEVE 



OXIDE COATING 



HEATER INSULATOR 



However, as stated, the emission from oxide-coated cathodes 
varies greatly, depending upon the core and coating materials, coat- 
ing and exhaust technic, etc. Furthermore, the life of cathodes 
depends also upon this technic and may vary greatly with it. 

PRACTICAL CATHODES 

Cathodes in vacuum tubes are universally heated electrically. 
The simplest type is in the form of a wire or ribbon, heated directly 

by passing a current through it. Vac- 
uum tubes having such filaments for 
cathodes are sometimes called fila- 
mentary tubes to distinguish them from 
tubes having indirectly heated cathodes. 
The filament current may be sup- 
plied by a battery, as shown in Fig. 1, 
or by a generator or transformer in 
case alternating current is employed. 
One disadvantage of using alternating 
current for the filaments of tubes used 
in audio-frequency circuits is that it 
introduces objectionable hum in the 
output. The hum can be minimized 
by connecting the plate and grid cir- 
cuits to the mid-point of the second- 
ary of the transformer, as shown in 
Fig. 4, but in general it is not possible 
to apply alternating current to the 
filaments of vacuum tubes used in the 
early stages of high-gain amplifiers. 

This difficulty is overcome in a large 
measure by using indirectly heated 

cathodes. One common arrangement is shown in Fig. 5. The cathode 
consists of a metallic cylindrical sleeve, usually of nickel, coated with 
a mixture of barium and strontium oxides. A lead- wire from the 
cathode sheath is carried out to an external tube terminal so that the 
cathode may be maintained at any desired potential. 

The heater wire is usually of tungsten, and may be in the form of a 
spiral or, as in the illustration, in the form of a hair-pin threaded through 
parallel tubular holes in a ceramic insulator. Vacuum tubes having 
cathodes of this type are referred to as heater type tubes ; or, since the 




HEATER WIRE 



FIG. 5. Schematic diagram of 
an indirectly heated cathode. 



Feb., 1935] 



THREE-ELECTRODE VACUUM TUBE 



143 



cathode sheath is at a uniform potential, they are often referred to as 
unipotential or equipotential cathode tubes to distinguish them from 
filamentary tubes. The heaters may be operated on either direct 
or alternating current. When the latter is used, although the hum 
is reduced to a much lower level than in filamentary tubes, it is not 
entirely eliminated, for reasons beyond the scope of this paper to 
discuss. The usual circuit arrangement is shown in Fig. 6. Con- 




FIG. 6. Schematic diagram of circuit connections to a triode 
with indirectly heated cathode. 




FIG. 7. 



Schematic diagram of circuit connections to a diode 
whose characteristics are shown in FIG. 8. 



necting the mid-point of the secondary winding of the transformer 
to the cathode usually reduces the residual hum observed even in 
heater type tubes. 



VACUUM-TUBE CHARACTERISTICS 



Let us next consider the characteristics of a vacuum tube con- 
taining a filamentary cathode and a plate or anode, as shown in Fig. 7. 
We shall assume that the filament current, I a , is held at some fixed 
value such that the filament operates at a temperature sufficiently 



144 



H. A. PlDGEON 



[J. S. M. P. E. 



high to emit a copious number of electrons. Let the plate be con- 
nected to the positive terminal of the plate battery, whose negative 
terminal is connected to the negative filament terminal as shown. 
We shall assume that the potential of this battery can be varied at 
will to any desired value. The potential of the plate, with respect 
to the negative filament terminal, is measured by the voltmeter, E P ; 
and the electron stream to the plate, called the plate current or space 
current, I P , is measured by the I P meter. Let us assume that the 




PLATE VOLTAGE, E, 



FIG. 8. Plate current-plate voltage characteristics of a diode. 



plate potential, E p , is varied gradually from zero to greater and 
greater positive values. The corresponding values of the plate 
current, I P , are measured by the I P meter. If the corresponding 
values of I P and E p are plotted, a curve is obtained as indicated by 
o, a, b, c, d in Fig. 8, showing that the current increases slowly at 
first as the plate potential is increased from zero, then more and more 
rapidly. Later it increases more slowly and finally becomes quite 
constant, increasing scarcely at all with further increase in plate 



Feb., 1935] THREE-ELECTRODE VACUUM TUBE 145 

voltage. The interpretation of the curve is as follows : At c the plate 
potential becomes sufficiently high to draw to the plate substantially 
all the electrons emitted by the filament and, consequently, there 
can be no appreciable further increase in current with further in- 
crease in plate voltage. In the region from c to d, then, the plate 
current is substantially independent of the plate voltage and the 
tube is in a condition of voltage saturation. The current is determined 
only by the temperature of the filament, designated by 7\. Now, 
at this point the question naturally arises: Since the plate, at any 
positive potential, exerts an attractive force upon all the negative 
electrons, why does it not when at some lower potential, such as at 
a, for example, attract all the electrons to it? The answer is that 
it does not because the evacuated space between the cathode and the 
plate offers resistance to the flow of the electron stream. The origin 
of this resistance is found in the presence of space-charge. 

Since the electrons are, in fact, minute negative charges of elec- 
tricity, there is at every point in the space between the cathode and 
the plate a negative charge whose density is proportional to the 
number of electrons per unit volume at that point. This charge is 
designated space-charge. Although it extends throughout the inter- 
electrode space, its density is relatively so much greater at points 
very near the cathode surface, where the velocity of the electrons is 
low, that for all practical purposes we may regard it as confined to a 
sheath about the cathode (in this case a filament) varying from a few 
thousandths to a few hundredths of a centimeter in thickness. The 
situation is somewhat as pictured roughly in Fig. 7. As a matter of 
fact, the space-charge is relatively much more dense at points close 
to the cathode than it is possible to indicate in such a picture. 

Now, this cloud of negative electrons produces an electrical field 
near the surface of the cathode which opposes the escape of electrons 
from it. This field is opposite in direction to that produced by the 
anode maintained at a positive potential. The resistance to the flow 
of electrons produced by the space-charge is such that if the cathode 
surface were at a uniform potential and if the electrons were emitted 
with zero velocity (which is not quite true), it can be shown from 
theoretical considerations that the plate current should increase 
proportionally to the three-halves power of the plate voltage. That is, 

/, = KE,*/* (2) 

in which K is a constant depending upon the geometrical dimensions 
of the electrodes and upon their spacing. The plate current, then, 



146 H. A. PlDGEON [J. S. M. P. E. 

from o to some point b, in Fig. 8 follows this three-halves power 
law, being limited by the resistance produced by the space-charge. 
In this region the plate current is independent of the temperature 
of the filament and, consequently, it is said to be in temperature 
saturation. 

If the temperature of the filament is lowered to some value T%> 
the plate current-plate voltage characteristic is changed as indicated 
by the curve o a b f c' d', the height of the flat portion c'd' again being 
determined by the lower temperature, T 2 , of the filament. 

Now, in the operation of vacuum tubes it is impracticable to main- 
tain the cathodes at a precisely constant temperature. Economic 
considerations require that certain tolerances be allowed in the range 
of operating filament current and temperature. On the other hand, 
it is highly undesirable to have the plate current and, consequently, 
the operating performance of the tube vary with such fluctuations 
in filament current. For this reason vacuum tubes must be limited 
in their range of operation to the region of reasonably good tempera- 
ture saturation. That is, in curve 1 of Fig. 8, the operating range is 
limited to the region o a b. If the temperature of the cathode in this 
particular tube should fall to T z , the characteristic would change from 
o a b to o a b' b" which would perhaps be too great a change to be 
tolerated. This subject will be discussed further after we have con- 
sidered the characteristics of three-electrode tubes. 

THE THREE-ELECTRODE VACUUM TUBE 

Suppose we modify the tube structure shown diagrammatically 
in Fig. 7, and whose characteristics are shown in Fig. 8, by inserting 
a grid between the plate and the cathode. Grids are usually spirals 
of fine wire held rigidly in place by welding them to heavier sup- 
porting wires. The modified structure typical of a three-electrode 
tube or triode is shown diagrammatically in Figs. 1 and 9. Let the 
filament and plate circuits be identical to those previously considered 
in Fig. 7. Let the grid be connected, as shown in Fig. 9, to the nega- 
tive terminal of the grid battery, the positive end of which is con- 
nected to the negative filament terminal. Let us assume that the 
grid voltage, E g , is variable at will from zero to any desired negative 
value, and that it can be measured by the voltmeter shown. 

Now, with the grid interposed between the cathode and the plate, 
we should expect it to shield the cathode from the plate more or less, 
and so render the latter less effective in drawing electrons away from 



Feb., 1935] 



THREE-ELECTRODE VACUUM TUBE 



147 



the cathode. Consequently, the plate current is smaller than before. 
Particularly, if the grid is held at some negative potential, it produces 
a negative field at the cathode which partially neutralizes the posi- 
tive field produced by the plate, thus reducing the power of the latter 
to pull electrons through the space-charge region. 

Let us first assume that the grid is held at zero potential with 
respect to the filament, and that the plate voltage is varied as before, 
reading the corresponding values of plate current, I P , for given values 
of E P . The results are represented by curve 1 of Fig. 10. 

Let us next assume that the grid is held at some fixed negative 




FIG. 9. Schematic diagram of circuit connections to a triode whose 
characteristics are shown in Figs. 10 and 11. 



potential, say, 3 volts, and the process repeated. The results are 
shown by curve 2. Let the process be repeated for E g = 6, 9, 
12, and 15 volts. The results are given by curves 3, 4, 5, and 6. 
We see that all the curves of this family of characteristics have 
approximately the same shape, and that they are approximately 
equally spaced. The approximate effect, then, of increasing the 
negative grid voltage by equal steps of 3 volts is to move the char- 
acteristic to the right by equal increments along the voltage axis. 
These curves will also be found to follow a three-halves power law 
approximately, if due regard is given to the transfer of the origin 
along the plate voltage axis. 

Now, let us assume that the plate is held at some fixed potential, 
say, 90 volts, and that values of the plate current are obtained as the 
grid bias is varied from zero to such a negative voltage that it com- 



148 



H. A. PlDGEON 



[J. S. M. P. E. 



pletely neutralizes the effect of the plate, thus reducing the plate 
current to zero. This characteristic is represented by curve 1 of 
Fig. 11. Let this process be repeated with the plate held at 120, 
150, 180, 210, and 240 volts. The results are given by curves 2, 3, 
4, 5, and 6. Again we obtain a family of similarly shaped curves 
spaced at approximately equal distances. Furthermore, the curves 
have approximately the same shape as the family shown in Fig. 10 



R-15DOO OHMS 



R-30,000 OHMS 



I I I 

R-8000 OHMS 



7 



Y 



A 



V 







120 160 

PLATE VOLTAGE , E, 



260 



FIG. 10. Plate current-plate voltage characteristics of a triode. 

and, in fact, they do follow the same law with certain necessary 
modifications. As an approximate equation for the entire family of 
characteristics of a three-electrode tube we may write, 

I p = K(E P + M J 3 /2 (3) 

in which /* is the amplification factor (discussed later) and K is a 
function of the geometrical dimensions of the tube elements and the 
amplification factor p. Both K and fj, are commonly referred to as 



Feb., 1935] 



THREE-ELECTRODE VACUUM TUBE 



149 



tube constants, although they do vary somewhat over the usual range 
of operation. In most cases, particularly in tubes having filamentary 
cathodes, an exponent differing somewhat from 3 /2 will be found to 
fit the characteristics best. Among the more important factors 
contributing to departure from the ideal three-halves power law 
are : potential drop along the filament, variations in the magnitude 
of the amplification factor, p, and the effect of the initial velocity 




-26 -24 -22 -20 -18 -16 -14 -12 -10 -8 -6 
GRID VOLTAGE. Eg 

FIG. 11. Plate current-grid voltage characteristics of a triode. 



with which the electrons are emitted from the filament. It is beyond 
the scope of this paper to discuss these factors in detail. 

It is readily seen from consideration of equation 3 that if K and ju 
were constant, all the curves of any family of characteristics, such 
as those shown in Fig. 10 or Fig. 11, would have exactly the same 
shape and would also have exactly the same horizontal voltage 
intercept between them at all points. For, if E g be changed by 
successive equal increments and if E P is changed at the same time in 
the opposite sense by successive equal increments /* times as large 



150 H. A. PlDGEON [J. S. M. P. E. 

as the increments inE g , the plate current, I P , will remain unchanged. 
This is true regardless of the value of I P . Hence, all the character- 
istic curves in the families shown in Fig. 10 or Fig. 11 could be ob- 
tained by a simple translation of any one of the curves by successive 
equal steps in the direction of the voltage axis. 

In both the families of characteristics shown in Figs. 10 and 11 
we have assumed that the total electron emission is sufficiently great 
that at all points the plate current is limited by the space charge. 
In other words, we are assuming that the total emission from the 
filament is at least several times the maximum values of the indicated 
plate current, so that the tube operates in a condition approximating 
temperature saturation. Consequently, the curves do not show any 
appreciable flattening out at the tops, such as shown in Fig. 8. 

From the families of characteristics shown in Figs. 10 and 11 it is 
apparent that under conditions of filament temperature saturation, 
the plate current in a triode is determined by both the plate and grid 
voltages. Since these voltages may be varied independently of each 
other, they may be chosen in pairs to determine an indefinitely large 
number of operating points extending throughout the permissible 
operating range of the tube. Each operating point (that is, each 
point in the diagram of Fig. 10 or Fig. 11) determines fixed values of 
plate voltage, grid voltage, and plate current. One such operating 
point is at P in Figs. 10 and 11, at which the plate voltage is 150 volts, 
the grid voltage is 9 volts, and the plate current is 5 milliamperes. 

Let us consider further the operating point P. If the grid voltage 
is varied about this point by the small amount, AE g (Fig. 11), indi- 
cated by the base of the small triangle abc, the plate current will at 
every instant be given by the corresponding points on the charac- 
teristic between the points a and c, and will vary through the range 
given by AI P which is the altitude of the small triangle abc. The 
ratio of this change in plate current to the change in grid voltage, 
with the plate voltage remaining constant, is called the mutual 
conductance or transconductance of the tube. That is, 

Transconductance = S m = ~ (4) 

AE<, 

when Alp and AE g are made infinitesimally small. 

Physically, the transconductance at any given operating point 
is the rate of change of plate current with variations in the grid 
voltage; that is, it is the slope of the plate current-grid voltage 



Feb., 1935] THREE-ELECTRODE VACUUM TUBE 151 

characteristic at constant plate voltage. It is usually measured in 
microamperes per volt or in micromhos. Numerically, it is the change 
of plate current in microamperes per volt change in grid voltage, the 
plate voltage remaining constant. 

Obviously, the transconductance is not the same at all operating 
points for a given tube. For example, if the operating point is chosen 
at P' (Fig. 11) and the same small change in voltage applied to the 
grid, the change in the plate current, A//, will be less than at P, 
because the characteristic curve is flatter at P' than at P. Conse- 
quently, the transconductance at P' is smaller than at P. At P" 
it is larger than at P for a similar reason. 

Transconductance varies greatly in different types of tubes. It is 
beyond the scope of this paper to discuss in detail the factors deter- 
mining its magnitude, but it will suffice to state that, in general, it 
increases with the size of tubes, with the plate voltage and plate 
current at which they operate, and also increases with decrease in 
the electrode spacings, particularly that between the cathode and 
grid. The transconductance is a useful criterion of performance, or 
figure of merit of vacuum tubes, since it determines the magnitude 
of output current for a given input voltage applied to the grid. 

Referring again to Fig. 11, suppose we start at the operating point 
P and change the grid voltage to E gZ with the plate potential held 
constant at 150 volts. This change from E gi to Eg is equal to the 
horizontal intercept PS, between the 150- and 120- volt character- 
istics. The plate current rises to the value given by the point Q on 
the 150-volt characteristic. Next, let the plate voltage be changed 
from 150 volts to 120 volts, maintaining the grid voltage constant 
at E g z. The operating point is now at S, and the plate current is 
exactly the same as it was at the original operating point P. We 
thus see that the effect upon plate current of reducing the plate 
voltage from 150 to 120 volts is exactly equal and opposite to that 
produced by changing the negative grid voltage from E g i = 9 
volts to E g 2 = 6 volts, because the two operations leave the current 
unchanged from its original value. The ratio of the plate voltage 
change to the grid voltage change producing equal and opposite 
changes in the plate current gives us the magnitude of another elec- 
trical parameter of the tube, known as the amplification factor, /*. 
Thus, in this case, 



152 H. A. PlDGEON [J. S. M. P. E. 

We see from this argument that to produce equal and opposite 
changes in the plate current (thus leaving its final value unchanged), 
the change in plate voltage must be /z times the change in grid voltage. 
In other words, we may describe the amplification factor as a measure 
of the relative effectiveness of small changes in grid voltage to similar 
changes in plate voltage in changing the plate current. The amplifica- 
tion factor is a very useful tube parameter since, in conjunction with 
the plate resistance, it at once gives a measure of the amplification 
to be attained with any given external load resistance. 

Let us next consider the operating point P in Fig. 10. It corre- 
sponds exactly to the operating point P in Fig, 11, since it defines 
the same plate voltage, grid voltage, and plate current. Now let the 
plate voltage be varied about this point by the small amount &E P , 
equal to the base of the small triangle abc. The operating point 
will simultaneously move along the characteristic from a to c, since 
the grid voltage is not changed. The corresponding change in the 
plate current is AI P which is equal to the height be of the small 
triangle abc. Now, since a change in plate voltage, &E P , produces a 
change in current, AI P , the quotient obtained by dividing &E P by 
Al p gives the resistance the tube offers to this change. This re- 
sistance is called the plate resistance, R P , of the tube at the operating 
point P. Thus, 



when AE P and AI P are made infinitesimally small. That is, R P 
is the reciprocal of the slope of the plate current-plate voltage char- 
acteristic at constant grid voltage. At the point P in Fig. 10, 

*' = 1.25 X 10-3 =8000 ohms 

By an argument precisely similar to that used in the case of the 
transconductance, it can be shown that the plate resistance at P' 
is larger than at P, and at P" it is smaller than at P. For all points 
along the line de, for which the current is constant, the plate resistance 
remains practically constant. 

We have just discussed three of the more important electrical 
parameters of the three-electrode vacuum tube, viz.: the trans- 
conductance, S m , the amplification factor, ju, and the plate resistance 
Rp. Both S m and R P vary widely with variations in plate current, 
and vary in an approximately inverse manner with respect to each 



Feb., 1935] THREE-ELECTRODE VACUUM TUBE 153 

other. The amplification factor, ju, is often referred to as the amplifi- 
cation constant and, while it is not a constant, it does not vary greatly 
over the usual operating range. 

These three tube parameters are not independent of each other 
and, in fact, we should not expect them to be, for the argument used 
in obtaining the value of ju indicates that some relation exists be- 
tween ju, S m , and R p . In fact, simply expressing the definition 
previously given for the amplification factor in mathematical form 

gives 

A/p 
AE, 



A/, 

AEp 



(7) 



If AEg and &E P be so chosen that I P remains constant that is, 
such that the values of A/^ in the numerator and denominator 
of equation (7) are equal in magnitude and opposite in sign then 



A P 1 
A < J/ P = 



(8) 
constant 



Equation (8) is equivalent to equation (5), and precisely defines the 
value of fj. when AE^ and AE a , are infinitesimally small. 
Combining equations (4) and (6) with (7) gives : 

M = SmR P 
from which 

&. = (9) 

In the illustration given above, 

s = 57^ = 125 X ID" 6 amperes per volt = 1250 micromhos 

5UUU 

at the point P. 

DYNAMIC CHARACTERISTICS 

Thus far we have considered only what are usually called the 
static characteristics of the three-electrode tube. That is, we have 
considered the plate current as a function of voltages applied di- 
rectly between the plate and cathode and between the grid and 
cathode without any external impedance between the measured po- 
tential sources and the tube elements. In the practical applications 
of vacuum tubes it is necessary to have such external impedances in 



154 



H. A. PlDGEON 



[J. S. M. P. E. 



the plate and grid circuits. For in any such application, the input 
signal is applied to the grid as a control element ; that is, the potential 
of the grid is made to vary with time in accordance with variations 
in the signal intensity. Since, as we have seen, the plate current 
varies with the grid potential, it also must vary in accordance with 
the signal input on the grid. Obviously, if the tube is to serve any 
useful purpose, its fluctuating plate current must be made to operate 
some receiving device such as a telephone receiver or a loud speaker, 
or to provide varying potentials to the grid of the following tube if 
an amplifier of more than one stage is employed. Since any such 
device must have impedance (that is, resistance), it necessarily re- 
sults in the introduction of an external impedance into the plate 




FIG. 12. Schematic diagram of circuit connections to a triode, 
with external load resistance and input resistance in the circuit. 



circuit of the tube. An exactly similar argument applies to the grid 
circuit. To apply an input signal to the grid requires the insertion 
of the input signal device in the grid circuit, with its given impe- 
dance. 

Let us next consider such an arrangement and the effect it has 
upon the operating characteristic of the tube. The circuit arrange- 
ment, in a simple form, is shown in Fig. 12. R is the resistance of the 
receiving device in the plate circuit. It is called the output or load 
resistance since it is the resistance into which the useful power is 
delivered. Ri is the impedance of the input signal device which 
applies an input voltage, e g , to the grid of the tube. Let us assume 
that we can vary this input voltage, e g , at will, and that for the present 
it is zero. We shall further assume that the grid is negative in 
potential with respect to the filament at all times so that it can not 
collect electrons. Since there is no current flowing in the grid circuit, 



Feb., 1935] 



THREE-ELECTRODE VACUUM TUBE 



155 



the potential drop across RI is zero, and thus the operating potential 
of the grid with respect to the filament is E C) which we shall assume 
to be 9 volts. This voltage is indicated by the voltmeter, E c . 

In the plate circuit, let the plate battery voltage E B be adjusted 
to give a potential between the plate and filament of 150 volts, as 
indicated by the voltmeter, E b . The plate and grid voltages are now 




-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 +2 
GRID VOLTAGE, Eg 

FIG. 13. Dynamic or load characteristics of a triode obtained 
with the circuit shown in Fig. 12, and with load resistances of 8000, 
15,000, and 30,000 ohms. 

adjusted to the values giving the operating point P in Fig. 11. Hence 
the operating plate current, I b , has the value 5 milliamperes deter- 
mined by that point. For the sake of clarity the curves of Fig. 1 1 
with the operating point P are reproduced in Fig. 13. 

It should be observed that with the operating plate potential, E b , 
adjusted to 150 volts, the plate battery voltage, E B , must be higher 



156 H. A. PlDGEON [J. S. M. P. E. 

than 150 volts. For, since the plate current, I b , flows through the 
load resistance, R, it produces a potential drop in the direction of the 
arrow.* This potential drop subtracted from the battery voltage 
E B , gives the operating potential difference E b between plate and fila- 
ment. That is 

E b = E B - hR (10) 

Starting at the operating point P in Fig. 13, let us vary the grid 
potential, maintaining the plate battery voltage, E B , constant, and 
see what happens. Since we have assumed that we can vary e g 
at will, let us adjust its value to 1.5 volts in such a direction as to 
reduce the grid bias from 9 to 7.5 volts. If the plate voltage 
were to remain constant at 150 volts, the operating point would be 
moved to P' on the 150-volt characteristic. The plate potential 
does not remain constant, however, for as the plate current increases 
the potential drop through R increases, and the plate potential falls 
an equal amount (since E B is constant) . Hence, the operating point 
must move to some point P% at a lower plate potential and lower 
plate current than at P 7 . Continuing this procedure by adding fur- 
ther positive increments to the input voltage, e g , we obtain additional 
points through which the curve P, P 2 , P 3 , P 4 is drawn. By choosing 
various negative values of e g , the curve is extended in the opposite 
direction from the original operating point P. This curve, PI, P, 
P 4 , is called the dynamic characteristic of the tube through the operat- 
ing point P. It is easy to determine the point of intersection of this 
curve with any static characteristic at constant plate voltage. For 
example, consider the point PS, at which it intersects the 120- volt 
characteristic. At this point the plate potential has dropped 30 
volts from its original value at P. This potential drop is caused by 
the increased potential drop across the load resistance, R, due to the 



* In the conventional sense, the current in the plate circuit flows from the 
positive terminal of the battery EB through the resistance R, thus producing a 
potential drop hR in the direction of the current flow. The current then flows 
from the plate to the filament and back to the negative terminal of the plate 
battery EB. Heretofore, we have referred to the electron current flow being 
from the filament to the plate. The reason for this seeming discrepancy is that 
the conventional direction of current flow is that which would be taken by a 
positive charge; whereas the electron, being an actual negative charge, moves 
in the opposite direction. 



Feb., 1935] THREE-ELECTRODE VACUUM TUBE 15? 

plate current increasing from I b to I P s, that is, by an increment, i^ 
We have then, assuming that R = 15,000 ohms. 

150 - 120 150 - 120 
= -- ^ -- = 15>OQO = 2.0milhamperes 

and I P z = h + 2*3 

= 5.0 + 2.0 = 7.0 milliamperes. 

Thus, the dynamic characteristic intersects the 120- volt characteristic 
at a plate current of 7 milliamperes. The dynamic characteristic 
may be defined as the locus of points determining the plate current 
of the tube as the grid potential varies, with a given load resistance 
in the plate circuit. 

The slope of the dynamic characteristic P\PP* at the point P 
may be derived readily as follows : Consider the two small triangles 
PP'b and PP^b in Fig. 13, with the common base AE g , and whose 
altitudes are given by A/^ and A//', respectively. These triangles 
are assumed to be so small that the static and dynamic character- 
istics are essentially linear throughout the range considered. The 
slope of the dynamic characteristic, which we shall call S md , is given by 



(ID 



Alp" = A/p - A7p' (12) 

We shall assume that the operating point moves from P to P 2 on the 
dynamic characteristic in two successive steps. In the first step the 
plate voltage remains constant and the grid voltage is changed by the 
amount AE g . The operating point moves from P to P' on the 150- 
volt static characteristic. The resulting change in plate current 
Alp, is given by 

A/p = S m AE g = -- AE a (13) 

RP 

In the second step, the grid voltage remains constant and the plate 
voltage is changed by the amount &E P , during which the operating 
point moves along the vertical line from P' to P 2 , the plate current 
changing by the amount A//. From the definition of plate re- 
sistance, E g , remaining constant, 



(14) 

p 

But this change in plate voltage is produced by the increased po- 



158 H. A. PlDGEON [J. S. M/P. E 

tential drop through the load resistance, R, due to the plate current 
increasing by the amount A/%; therefore, 

AEp = Alp'R (15) 

Hence, 

A/,' = - A// (16) 

-KP 

Substituting the values of A/^, and A// given by equations (13) 
and (16) in (12) 

A// = -jj- AE, - ^- A V 
or 

A r H V A IT 



Finally, substituting in equation (11) 



M 



i p 
-\- K 



(1) 



Equation (18) expresses the slope of the straight line cd in Fig. 13, 
which is tangent to the dynamic characteristic at the operating 
point P. If the dynamic characteristic were linear, that is, if it coin- 
cided with the straight line cd at all points, the output current, i* 3 , at 
any point P 3 would be exactly proportional to the input voltage, e gZ , 

* To those familiar with calculus operations, it will be apparent that the above 
procedure is equivalent to a simple process in differential calculus. 
For, let 

I p =/(p, ) (19) 

Then 



Which, from the definitions of transconductance and plate resistance, may be 
written, 

dl p = ^- dE p + -- dE g (21) 

J\.p K p 

Also for points on the dynamic characteristic, 

dE p = - dlp-R (22) 

Substituting in the preceding equation, 

*/,-- dip + |- dE. (23) 

From which 

5 ^ = p ^ /24) 



Feb., 1935] THREE-ELECTRODE VACUUM TUBE 159 

and its value would at once be obtained by multiplying equation (18) 
by the input voltage, e g $. Thus, 

i^ = -s-T-5 ' ** (25) 



For any signal input, the output current and the resulting varia- 
tion in potential across the load resistance, R, would be an exact rep- 
lica of the signal voltage in magnified form. But as the dynamic 
characteristic is not exactly linear, the output is not quite propor- 
tional to the input. There results what is known as distortion in the 
output, a subject that will be discussed more fully later. 

For very small inputs, the dynamic characteristic practically 
coincides with the straight line cd and, consequently, the output 
current is given quite accurately by equation (25). For larger signal 
inputs, it can be shown that the fundamental components of the 
output, that is, that portion of the output that is a replica of the 
input, is given to a fairly close approximation by equation (25) . 

As a further illustration of the foregoing argument, let us assume 
that a very simple signal, which varies with time, is applied to the 
grid. We shall assume that, before this signal is applied, the plate 
voltage impressed upon the tube as measured by the voltmeter, E b , is 
150 volts, and the grid bias voltage as measured by the voltmeter, E c 
(Fig. 12), is 9 volts. The operating point of the tube is now given, 
as before, by the point P in Fig. 13. To avoid confusion, these curves 
are reproduced in Fig. 14. Now, let us assume that an input voltage 
which is sinusoidal in form is applied in the grid circuit. The form 
of such a voltage wave is shown by curve A in Fig. 14. Time is 
measured downward from the point on the vertical axis through 
the point P. The input voltage at any time T\ is given by the hori- 
zontal displacement of the corresponding point Q\ on the curve. 
It will be seen that as time progresses from O, the input voltage in- 
creases toward the right or positive direction until at time T 2 it 
reaches a maximum value, e g , at Qz\ then recedes to zero at Tz\ 
increases to an equal maximum distance, e g , in the negative direction 
(left) ; and again returns to zero at time T 5 . The voltage is assumed 
to repeat this cycle continuously in periods of time equal to T 5 . 

Let us now inquire what is the character of the output of this tube 
with such an input voltage applied. It is given by curve B(0'Q2 r Qt f ), 
in which time is now measured in the horizontal direction toward 
the right from the point 0', and the current is given by the ordinates . 



160 



H. A. PlDGEON 



[J. S. M. P. E. 



This may be readily verified point by point. For example, consider 
point Qi on curve A . Projecting this point vertically upward shows 
that it corresponds to a grid voltage of 2.6 volts and to a plate 
current of 7.86 milliamperes at point <2i" on the dynamic character- 
istic. Projecting this point horizontally to the right gives the point 
Qi> at time r b on curve B. Other points on curve B may be obtained 
from curve A in the same manner. 




FIG. 14. Output current of a triode with a sinusoidal input voltage applied 

to the grid. 

Curve B, showing the variation of the output current about the 
time axis O'T', is almost, although not quite, exactly similar to the 
input voltage curve A. The positive lobes above the axis O'T' are 
somewhat higher than those below this axis. This distortion is due 
to the curvature of the dynamic characteristic as previously dis- 
cussed. If the dynamic characteristic were the straight line cPd, 
then the output current would be exactly given by the dashed curve 
O'Qz'Qt", and the output current would also be sinusoidal and of 
exactly the same form as the input wave. 

From equation (18) it is apparent that the slope of the dynamic 
characteristic decreases with increasing values of the load resistance, 



Feb., 1935] THREE-ELECTRODE VACUUM TUBE 161 

R. If the load resistance is made equal to the plate resistance at 
point P in Fig. 13, that is, R = 8000 ohms, the dynamic characteristic 
is given by curve 1. If the load resistance is changed to 30,000 ohms, 
the dynamic characteristic is given by curve 3. Increasing the 
load resistance not only decreases the slope of the dynamic char- 
acteristic but it also decreases the curvature, making the character- 
istic more nearly linear. This reduces the distortion, although at 
the same time it reduces the power output when R becomes larger 
than R P , since it can be readily shown that for a given grid swing, the 
maximum power is obtained from a vacuum tube when the external 
load resistance is made equal to the plate resistance. 

In vacuum tubes associated with telephone transmission lines, it is 
desirable or even necessary to operate them into matched impedances 
in order to minimize disturbing effects due to reflection. However, 
in many applications where this restriction does not apply and where 
distortion is a criterion of the useful power obtainable, it is advan- 
tageous to operate a three-electrode tube into a load resistance 
larger than the plate resistance, at the same time choosing an opti- 
mum point on the plate current-plate voltage characteristic such as 
to give the maximum permissible plate-voltage and plate-current 
swings. This will be illustrated after we have discussed a simple 
method for the approximate computation of the percentage of dis- 
tortion. 

It is sometimes advantageous to work with dynamic character- 
istics plotted on the plate current-plate voltage family of character- 
istics. Dynamic characteristics, commonly called load lines, corre- 
sponding to curves 1, 2, and 3 of Fig. 13, are shown in Fig. 10. Each 
dynamic characteristic or load line so plotted is a straight line whose 
slope is equal to^the reciprocal of the load resistance. This follows 
from the fact that at all points on any dynamic characteristic the 
variations in plate potential are equal to the variations in the po- 
tential drop across the load resistance, which in turn are directly 
proportional to the variations in the plate current. 

AMPLIFICATION 

The final objective in an amplifier tube is, of course, not only to 
have the output as nearly identical to the input in wave-form as 
possible, but also to have it as much larger as possible as measured 
by its current, voltage, or power. 

As an illustration, let us compute the amplification obtained in the 



162 H. A. PIDGEON [J. S. M. P. E. 

example previously considered with a sinusoidal input applied to the 
grid. The amplitude or peak value of the input voltage is given by 
the line T 2 Qz in curve A of Fig. 14, which we shall call e g . For rea- 
sons previously given, we may take as a close approximation the 
ordinate Toft" of curve B, which we shall call i p , as the peak value 
of that component of the putput which is an exact replica of the 
input wave. Then i p is at once obtained by multiplying the dynamic 
transconductance, given by equation (18), by e g . That is, 

* - e (26) 



The peak value of the output voltage wave, e p , across the load re- 
sistance is obtained by multiplying i p by R. Whence, 

e * = l^r-i, R (27) 



The voltage amplification, A v , is given by the ratio of e p to e g ; that is, 

A - = I - dh ' (28) 

In this case, R = 15,000 ohms, R p = 8000 ohms, and p, = 10. 
Hence, 

15,000 _ 15 

~ 8000 + 15,000 M ~ 23 M : 

Thus, in this simple case the output voltage is 6.52 times the input 
voltage. 

The power amplification is also readily computed. The average 
power dissipated in the load resistance, with a sinusoidal current 
flowing through it, is given by one-half the product of the peak 
current by the peak voltage across its terminals. That is, 



._ i y e _ 

~ lp ' 6p ~ 



2 p ' p ~ 2 R p +R R p +R 



e R - 1 ( /* 6 <> \ 2 

+R K ~ 2 \R P + R) 



In the input circuit the grid draws no current, as previously dis- 
cussed, so that the only power dissipated is in the resistance, R it 
due to the input voltage, e g , across it. The peak current through R t 
is given by 

- 1 (30 > 



Feb., 1935] THREE -ELECTRODE VACUUM TUBE 163 

The power dissipated in R^ is given by one-half the product of the 
peak current and peak potential drop that is, 

W -* V P - C\\\ 

yy, . n*0 * 6 o p \3L) 

& & j\i 

The power amplification is then given by 



= = ' R Ri (32) 



If Ri = 600,000 ohms, 



100 
A u = (80004.15000)8 ' X i 5 ' 000 x 600,000 = 1701-fold amplification 

In many applications it is convenient to express the amplification 
in logarithmic form. It is commonly expressed in terms of ten times 
the logarithm to the base ten of the power amplification. The 
amplification expressed in this manner is called the gain of the tube, 
and the unit of measurement is the decibel, indicated by db. Thus, 
gain in the above example is given by 

G = 10 logio (33) 



(34) 

(35) 



32.3 db. 



Gain values recorded for Western Electric vacuum tubes are so 
measured that they represent the values .computed by equation (35) 
when Ri = 600,000 ohms, and R is fixed at a value approximately 
equal to the average value of R p for the given type of tube, under the 
voltage conditions at which the measurements are made. 

APPROXIMATE METHOD FOR COMPUTING DISTORTION 

We have seen that because the static characteristics of the three - 
electrode vacuum tube are curvilinear in form, the dynamic char- 
acteristic must also be non-linear. The curvature of the dynamic 
characteristic is decreased by increasing the load resistance, approach- 
ing linearity when the load resistance becomes very large compared 
to the plate resistance of the tube. This non-linearity results in 
distortion in the output of the tube. That is, for any given wave- 
form applied to the grid, the amplitudes of the output current or 



164 



H. A. PlDGEON 



[J. S. M. P. E. 



voltage components are not exactly proportional to the amplitudes 
of the same components in the input; furthermore, additional fre- 
quency components appear in the output, which are harmonics of 
the input frequencies or the products of cross-modulation between 
them. 

If the form of the dynamic characteristic is known, it is possible 
to compute the magnitude of the various distortion components in 
the output with a precision depending only upon the degree of ac- 
curacy with which the dynamic characteristic is known. In the case 




-13 -12 -11 -10 -9 -8 -7 -6 -5 -4 
GRID VOLTAGE.Eg 

FIG. 15. Dynamic load characteristic of a triode. 



of load impedances having reactive components, the computations 
are rather involved; but in the case of pure resistance loads the 
computation is relatively simple. 

Simple equations for the computations of the approximate power 
output and distortion for three-electrode tubes with resistance loads 
have been given in the literature, 1 and have been in common use for 
several years. By making certain simplifying assumptions as to the 
form of the dynamic characteristic, these equations are easily de- 
rived. 

Let us assume that curve P\PPz of Fig. 15 is the dynamic char- 
acteristic of a three-electrode tube, obtained either experimentally 



Feb., 1935] THREE-ELECTRODE VACUUM TUBE 165 

or by computation from the static characteristics, in the manner 
previously given. Let us further assume that, taking the point 
as origin, curve PiPP% is represented by the equation 

I P = I b + aiv + a^ (36) 

in which a\ and a 2 are constants; I b is the steady value of the plate 
current with plate voltage, E b , and grid voltage, E c , applied to the 
tube and with no variable input applied to the grid ; v is the variable 
voltage applied to the grid, measured from the axis PO, and which 
in this case we shall assume to be sinusoidal in form and given by 

v = e a sin ut (37) 

aw is, obviously, the equation of the straight line APB with the point 
P as origin, and whose slope is a\\ a 2 v 2 is the equation of the curve 
PiPPz with P as origin, and its ordinates measured from the line 
APB. 

Ip attains its maximum value, I max ., and its minimum value, 7 min , 
when v assumes its maximum positive value, + g , and maximum 
negative value, e g , respectively. Substituting v = +e g and v = 
e g in equation (36), the following equations are obtained: 

I max . = Jb + aie g + a* Q * (38) 

Imin. = Ib a\e a + atfg* (39) 
Subtracting and adding equations (38) and (39) 

aie g = V (/.,. - J m -O (40) 

(hfi g * = l /2 (Ima x . + Imin. ~ 2/ 6 ) (41) 

Now, if we substitute the assumed value of v from equation (37) 
in equation (36), we obtain 

I p = I b -j- a\e g sin tat + a 2 e g z sin 2 wt (42) 

which reduces to 

IP = /& + l /t <hP * +a\e a sin co/ 1 / 2 a& a * cos 2wt (43) 

We see that, with the simple form of dynamic characteristic assumed, 
the operating plate current I b is increased by the non-periodic compo- 
nent 1 /2 #20g 2 ; and the total steady component of plate current is 
represented by the ordinate OC in Fig. 15 The term a\e g sin co/ 
is the amplified output given by the ordinates of the straight line 
CiCCz measured from the axis C\CC* . As previously shown 

* - (44) 



166 H. A. PIDGEON [J. S. M. P. E. 

In addition, there is in the output the second harmonic distortion 
term, l /z Ovtg 2 cos 2co/, whose magnitude with varying values of co/ 
is given by the ordinates of the curve P\PP<i measured from the line 
dCC 2 . 
The output power of fundamental frequency is given by 

W = Vf <*iV* (45) 

which, on substitution of the value of a^ from equation (44), is identi- 
cal to equation (29). 

Substituting the value of a\e g from equation (40) in equation (45) 
gives 

W = V (/m,. ~ Imin^R (46) 

The maximum potential variation across the load resistance is 
equal to the maximum difference in plate voltage, and is given by 

2aie a R = (I max . - I min )R = E max - E min (47) 

Substituting in equation (46), 

W Q = V8 (/max. - /*.) (., - E min ) (48) 



We may take as a measure of the distortion the ratio of the ampli- 
tude of the second harmonic to the amplitude of the fundamental. 
Calling this ratio D, we have 



Substituting in equation (49) the values of a\e g and a 2 e g 2 from equa- 
tions (40) and (41). 

Daa 1W +/,<- 2/ (5Q) 

& -Lmax -*mtn. 

Ordinarily it is not possible to express the dynamic characteristic 
of a three-electrode tube precisely by an equation as simple as equa- 
tion (36). In general, the addition of third and higher power terms 
in v is necessary. This results in the appearance of third and higher 
order harmonic tejms in the output. Consequently, equations (46), 
(48), and (50), must be regarded only as somewhat rough approxi- 
mations. 

In Table I, the power output and percentage of distortion com- 
puted by equations (46) and (50) are tabulated for the four dynamic 
characteristics shown in Fig. 13. With a plate potential of 150 volts 
and a grid potential of 9 volts, the plate resistance is 8000 ohms. 
When the tube works into a load resistance matching this resistance, 



Feb., 1935] THREE-ELECTRODE VACUUM TUBE 167 

the output power is seen to be 134 milliwatts with 12 per cent dis- 
tortion. As the load resistance is increased, the output power de- 
creases and also the percentage of distortion. With a load resistance 
of 15,000 ohms the output is 107 milliwatts with 3 per cent dis- 
tortion. If the tube is required to work into a resistance matching 
its own plate resistance and with the distortion limited to 3 per cent, 
the maximum power of this quality is obtained with a grid bias of 
6.7 volts, at which point the plate resistance and load resistance 
are each 6800 ohms. The output power is 80 milliwatts, or 75 per 
cent of the power of the same quality attainable at a grid bias of 9 
volts and with a load resistance of 15,000 ohms. Furthermore, the 
plate current in the latter case is only 5.0 milliamperes compared 
to 8.1 milliamperes in the former. 

TABLE I 

Plate Potential = 150 Volts 



Operating 
Point, 
Fig. 13 


EC 
(Volts) 


ib 

(Milli- 
amperes) 


RP 
(Ohms) 


R 
(Ohms) 


eg 
(Peak 
Volts) 


Wo 
(Milli- 
watts) 


D 
Per 
Cent 


P 


-9 


5.0 


8,000 


8,000 


9 


134 


12 


P 


-9 


5.0 


8,000 


15,000 


9 


107 


3 


P 


-9 


5.0 


8,000 


30,000 


9 


83 


1.2 


P" 


-6.7 


8.1 


6,800 


6,800 


6.7 


80 


3 



TEMPERATURE SATTTRATION AND FILAMENT ACTIVITY TESTS 

We shall now return to the subject of temperature saturation 
or filament activity, and the effect of insufficient activity upon the 
operation of vacuum tubes. In the foregoing discussion of the three- 
electrode tube as an amplifier, we have assumed that the electron 
emission from the cathode is sufficiently large so that the static 
characteristics remain essentially unchanged over the range of fila- 
ment current and temperature fluctuations permitted in service. 

Now, it would be very difficult and uneconomical to maintain 
the filament current and temperature constant in practice. The 
amount of variation that must be tolerated varies with the service. 
For example, in the telephone plant a variation of plus or minus 
about 3 per cent from the mean filament current is allowed. In 
applications where small local batteries are employed, a wider range 
of variation must usually be tolerated. Where commercially sup- 
plied power is employed, a wider range of plus or minus 5 per cent 



168 



H. A. PlDGEON 



[J. S. M. P. E. 



and more must usually be tolerated, depending upon the locality 
and the type of service. 

Let us assume that we have a tube sufficiently deficient in electron 
emission from the cathode so that the static characteristics vary 
appreciably from time to time with more or less erratic variations 
in filament current, over the permissible range. It was shown in the 



-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 +2 




FIG. 16. Plate current-grid voltage characteristics of a triode, showing 
the effect of low electron emission. 



foregoing discussion that the electrical parameters such as mutual 
conductance and plate resistance, and the amplification as well, de- 
pend upon the shape of the static characteristics. Consequently, 
these factors must change if the static characteristics change due to 
inadequate electron emission. 

This point is illustrated by the curves of Fig. 16. The curves 
drawn in solid lines may be taken here to represent the static char- 
acteristics of a tube of somewhat low filament activity when its 



Feb., 1935] THREE-ELECTRODE VACUUM TUBE 169 

filament current is at the highest point in its permissible range. 
The curves in dashed lines represent the static characteristics of the 
same tube when the filament current and temperature drop to the 
lowest values permitted. The plate current is seen to be lower, the 
difference being proportionally greater at the higher values of plate 
current. Assuming that the plate battery voltage, E B , remains 
unchanged in the circuit shown in Fig. 12, the operating point is now 
P' instead of P. The slope of the characteristic at point P' in 
Fig. 16 is smaller than at P. Consequently, the transconductance 
is decreased from the value at P. In a similar manner, it may be 
shown that the plate resistance is correspondingly higher. Further- 
more, the slope of the dynamic characteristic or load curve is de- 
creased. From equations (26), (29), and (35) it is clear that this 
must result in a reduction in the output current, output power, and 
also in the gain. 

Not only does a tube of low filament emission show much greater 
variations in these factors with filament current variations than 
does a tube with high emission, but it also usually shows lower 
absolute values of plate current and gain and a higher value of 
plate resistance at the normal filament current than the same tube 
would show were the filament emission higher. This is evident from 
the curves of Fig. 16, but is further illustrated by the curves of Fig. 
17. These curves show the gain of vacuum tubes plotted against 
filament current expressed in per cent of its normal value. The 
normal operating current is indicated by the vertical line aa' . Curves 
1 and 2 are for good tubes of high filament emission, one with 
average gain and the other with the lowest gain permitted by the 
manufacturing test specification. The difference in gain between 
the two tubes is due to a difference in structural dimensions. In 
both cases the gain is seen to remain practically constant when the 
filament current is lowered to the value indicated by the line W '. 
Curve 3 is for a tube of medium emission, which has a value of gain 
at aa' intermediate between curve 1 and curve 2 but which falls 
off more rapidly when the current is lowered to bb ' . Curve 4 is for a 
tube of low emission. Not only is the gain appreciably lower at aa' 
than if the filament were in good saturation, but it also falls off 
much more rapidly as the filament current is reduced tobb'. 

Not only does the thermionic emission vary markedly in different 
tubes of the same type, but it also varies with life in any individual 
tube. While the thermionic history of vacuum tubes differs widely, 



170 



H. A. PlDGEON 



[J. S. M. P. E. 



we may illustrate a typical life cycle approximately by curves l t 3, 
and 4 of Fig. 17. The initial state is illustrated by curve 1. It is 
seen to remain very flat until the filament current falls far below its 
normal operating value, when the gain falls off very rapidly. At 
a much later period in the life of the tube the condition is represented 
by curve 3. The curve at the operating current aa' is not as flat as 
curve 1, the knee of the curve has approached much nearer to the 
operating current, and, consequently, the filament is not nearly so 
far in temperature saturation as at first. The tube is assumed still 
to operate satisfactorily at this time. The time elapsing between the 
conditions represented by curve 1 and curve 3 represents by far the 



29 



7 



" 



70 75 8O 85 90 95 100 105 1K> 115 120 

FILAMENT CURRENT-PER CENT OF NORMAL VALUE 

FIG. 17. Gain of vacuum tubes as a function of filament current, show- 
ing the effect of low electron emission. 



greater portion of the life of the tube. From this point the deteriora- 
tion increases at a more rapid rate until, in a period relatively short 
compared to the previous period, the condition is represented by 
curve 4. Here the gain has dropped to a lower value, and changes 
so rapidly with decreasing filament current that the tube has become 
so unstable as to be unsatisfactory for further use, and must be re- 
jected. 

If, for the same tube, mutual conductance or plate current were 
plotted against filament current, curves similar in shape would be 
obtained. If the plate resistance were plotted, curves would be 
obtained rising rapidly instead of falling with decreasing filament 



Feb., 1935] THREE-ELECTRODE VACUUM TUBE 171 

current, because the plate resistance increases as the filament falls 
out of temperature saturation. 

It is highly desirable to have some means of testing vacuum tubes 
to determine when their thermionic emission has decreased to a point 
where they can no longer satisfactorily perform their required function 
or have become so unstable as to constitute a hazard to the service. 
Such a test is called a filament or cathode activity test, the limits 
for which will depend upon the character of the service in which the 
tubes are employed. The most satisfactory test of this sort is to 
measure some electrical parameter or property of the tube such 
as the transconductance, plate current, plate resistance, or the gain, 
at the normal operating filament current or voltage represented by 
the line aa r in Fig. 17, then to repeat the measurement at a reduced 
filament current or voltage corresponding to the line bb f . Both the 
absolute value of the quantity measured and the percentage change 
may be taken into consideration in determining the quality. 

CIRCUIT CONNECTIONS 

In practice, a variety of methods are employed for connecting 
vacuum tubes in the circuits with which they are associated. A 
common method is that shown in Fig. 18, where the vacuum tube 
is connected to the input and output circuits by means of trans- 
formers TI and T%, respectively. Transformer coupling is widely 
used in telephone amplifiers or repeaters, in the final or output stage 
of radio receivers, and in the output stages of amplifiers used in 
public address systems, sound-picture amplifiers, etc. The input 
terminals 1 and 2 of the input transformer, TI, are connected to the 
source of the signal it is desired to amplify, such as a transmitter, or 
the output terminals of a preceding vacuum tube in case an amplifier 
of more than one stage is used. The output terminals 3 and 4 of the 
output transformer, Tz, are connected to the output load which, for 
example, may be a loud speaker or the input terminals of a following 
vacuum tube. The grid circuit and plate circuit are usually con- 
nected to the negative filament terminal as shown, and this point 
of connection, c, is referred to as the common point. Transformer 
coupling, especially for power amplifier tubes, has several advantages, 
among which are: (1) By proper design the transformers can be 
made to match the impedances between which they work. This 
results in high transmission efficiency; that is, it makes the best 
possible use of the input power available at terminals 1 and 2 to obtain 



172 



H. A. PlDGEON 



[J. S. M. P. E. 



the largest power possible under the conditions imposed, at the out- 
put terminals 3 and 4. (2) By making the number of turns of wire 
in the secondary, Si, of the input transformer larger than the number 
in the primary, P,, a step-up is obtained in the voltage available to 
apply to the grid of the tube. This, obviously, increases the voltage 
amplification obtained. (3) The primary, P 2 , of the output trans- 
former, T 2 , can usually be so designed as to have a reasonably low 




FIG. 18. Schematic diagram of a triode with transformer coupling. 




FIG. 19. Schematic diagram of a triode with the plate current 
supplied through a retard coil and with transformer coupling. 



d-c resistance and, consequently, the potential drop due to the 
operating plate current, I b , flowing through it is small. Hence the 
actual steady potential of the plate, E b , is almost as large as the battery 
voltage, E B , instead of being much smaller, as is the case in the 
circuit shown in Fig. 12. On the other hand, transformer coupling 
has the disadvantage that amplifiers so connected amplify uni- 
formly over only a relatively limited range of frequencies. Equali- 
zation or another type of coupling is necessary where uniform ampli- 
fication is necessary over very wide frequency ranges. 



Feb., 1935] 



THREE-ELECTRODE VACUUM TUBE 



173 



In some cases it is not desirable to have the plate current, I b , flow 
through the primary winding, P 2 , of the output transformer as shown 
in Fig. 18. A means of avoiding this, but still retaining transformer 
coupling, is shown in Fig. 19. The plate current, I b , is supplied to the 
plate through the retard coil, r. A retard coil has very low resistance 
to direct current but very high impedance to alternating current. 
Consequently, the plate voltage, E b , is very nearly equal to the plate 




! E B 




FIG. 20. Schematic diagram of a triode with resistance coupling. 




I 




FIG. 21. Schematic diagram of modified resistance coupling, in 
which the plate battery is connected to the plate through a retard 
coil. 

battery voltage, E B . The alternating-current components produced 
by the signal are almost completely excluded from this branch of 
the circuit, due to the high impedance of the retard coil for such 
currents, and are constrained to flow through the condenser, C, and 
the transformer primary, P 2 . The so-called blocking condenser, C, 
prevents direct current from the battery, E B , from flowing through 
this path. 

Where it is desired to amplify uniformly signals covering a large 
range of frequencies, resistance coupling is often employed. A typical 



174 H A. PlDGEON 

circuit arrangement for this type of coupling is shown in Fig. 20. 
The plate current from the battery, E B , flows through the resistance 
Ri t which is usually a large resistance of, say, 20,000 to 100,000 ohms, 
or even higher. The potential drop across R\ is high and, conse- 
quently, the plate battery voltage, E B , must be very much larger than 
the operating plate voltage, E^. R% is usually large, in the range from 
one hundred thousand to several hundred thousand ohms. By 
proper choice of R\ t R*, and C, resistance-coupled amplifiers can be 
designed to amplify over a large range of frequencies with a high 
degree of uniformity. Since the resistance, RI, must be large, this 
type of coupling is best adapted to small tubes of low plate current 
drain, used in the early stages of high-gain amplifiers. 

Where the requirements as to uniformity of amplification with 
frequency are somewhat less stringent, resistance coupling may be 
modified as shown in Fig. 21 where a retard coil, r, replaces the re- 
sistance, RI, of Fig. 20. As previously discussed, the d-c. resistance of 
the retard coil is comparatively small, and E B need be only slightly 
larger than the required plate potential. Rz may be arranged as 
shown in Fig. 20 or in the form of a potentiometer, as shown in Fig. 
21, which permits variation of gain by varying the input voltage 
applied to the grid of the tube in the next stage. 

REFERENCES 

1 BROWN, W. J.: "Discussion on Loud Speakers for Wireless and Other Pur- 
poses," Proc. Phys. Soc. (London), 36 (Apr. 15, 1924), Part 3, p. 218. 

KELLOGG, E. W.: "Design of Non-Distorting Power Amplifiers," J. A. I. 
E. E., 44 (May, 1925), No. 5, p. 490. 

WARNER, J. C., AND LOUGHREN, A. V.: "The Output Characteristics of Amp- 
lifier Tubes," Proc. I. R E., 14 (Dec., 1926), No. 12, p. 735. 



THE 16-MM. SOUND-FILM OUTLOOK* 
W. B. COOK** 



Summary. After pointing out that the original difficulties in developing 16-mm. 
sound-film seemed only to spur the engineers to accomplish their solutions, it is shown 
how today a program of eight full reels of sound-film can be projected with quality ex- 
ceeding that of 35-mm. quality at a corresponding stage of its development. The short- 
age of sound-film subjects that hampered the distribution of 16-mm. equipment is 
rapidly being relieved, and interest in the industrial and institutional uses of sound 
film are growing. Future progress depends upon the equipment manufacturers; 
the film problems have been solved. Projectors must be made simple to operate, the 
frequency response must be broadened, they must operate silently; and, most impor- 
tant of all, their ultimate success will depend upon the cost of equipment and film ser- 
vice. 

Older members of the Society will recall that in the early stages of 
sound recording it was decided that the former camera and projector 
standard speed of sixty feet a minute would be inadequate for re- 
cording sound, and the standard speed was arbitrarily changed to 
90 feet a minute. Even with the greatly extended length of available 
recording track that resulted, several years of the most intensive re- 
search and experimental work were required to achieve uniform re- 
sults of even passable quality. 

To have predicted at that early date that a satisfactory sound 
track could be properly recorded upon and reproduced from a strip 
less than one-sixteenth of an inch wide and with a projection speed 
of only 36 feet per minute, would have seemed a wild hope or an idle 
dream. But the difficulties to be overcome served as a spur, and some 
of the cleverest engineers of the research laboratories turned at a 
comparatively early date to solving the difficult problems involved 
in producing and reproducing satisfactory 16-mm. sound-film. 

During the earlier stages of research and experiment, the publicized 
results seemed to favor re-recording from 35-mm. to 16-mm. sound 
tracks as a means of making 16-mm. sound-films, but the Kodak Re- 
search Laboratories felt that photographic reproduction by optical 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 
** Kodascope Libraries, New York, N. Y. 

175 



176 W. B. COOK [j. s. M. p. E. 

printing offered the more promising results. Such has proved to be so, 
and optical reduction is now the only system employed commercially. 

Inasmuch as the decrease of length of the sound-track was 60 per 
cent, whereas the decrease of width was less than 15 per cent, optical 
reduction presents rather complicated problems, which have been 
met by different laboratories in several different ways, with more or 
less success. It is not the purpose of this paper to enter into the 
technical details of sound reproduction, but merely to state that a 
number of laboratories are now turning out 16-mm. sound prints 
from 35-mm. negatives of a quality distinctly superior to that of 
35-mm. theatrical pictures at the corresponding stage of development. 
It should be remembered that dimensions of the photographic image 
of a 9000-cycle recording on 16-mm. film (which is easily attainable 
by optical reduction) would be about the same as those of a 22,500- 
cycle recording on 35-mm. film, if such a frequency could be audibly 
produced and recorded. 

At this point, it may be in order to point out the efforts being made 
by European countries, particularly Germany, to adopt a standard of 
16-mm. sound-film that shall have the same dimensions and relations 
of the picture image and the sound-track as those adopted by this 
Society, but exactly reversed as to the projection positions of the 
emulsion surface and the perforations. In the standard adopted by 
the Society of Motion Picture Engineers, when the film is threaded 
in a projector the perforations are at the right edge of the film (look- 
ing toward the screen), and the emulsion surface is toward the screen. 
In the standard advocated by European societies, the perforations 
are at the left edge of the film (facing the screen), and the emulsion 
surface is adjacent to the light and the condenser lens. The com- 
plication is clearly recognized by at least one European manufacturer, 
who advertises that he will supply either right-hand or left-hand 
sprockets as required. 

Considering the development of sound-film projectors, it is gratify- 
ing to observe the progressive spirit that has inspired manufacturers 
in the field. All the leading manufacturers of 16-mm silent pro- 
jectors have already produced or are now working on sound pro- 
jectors also. The RCA Victor Company has been the real pioneer in 
manufacturing 16-mm. sound-film projectors, and is already exploit- 
ing its third model. Thus a comparatively wide choice of projection 
equipment of varying capacity and price range is available. Most 
of the sound projectors are adapted to both 400- and 1600-foot reels, 



Feb., 1935] 16-MM. SOUND-FILM OUTLOOK 177 

the latter permitting a program of eight full reels to be projected, with 
a single interruption for rewinding. As many theatrical features are 
only six or seven reels in actual footage, it becomes practicable to 
precede the feature by one or two shorts and yet have the entire pro- 
gram on two large reels. 

A decidedly interesting recent development for the amateur has 
been a portable and compact 16-mm. sound-film camera, with which 
the amateur can make his motion pictures with a sound-record of the 
operator's voice. Accessories for recording the voice and sound- 
effects made by the subjects photographed are available at addi- 
tional expense, but are for the present rather heavy and bulky. 
With further research and competitive manufacture and production, 
however, it is reasonable to expect that the amateur sound-and- 
picture camera will soon become practicable and popular. 

Until recently, the sale and distribution of sound projectors have, 
been hampered principally by the lack of an adequate supply of 
available sound-film entertainment subjects. Happily, this shortage 
is now being very rapidly remedied, and perhaps the most outstanding 
development in the new field has been the astonishing increase in the 
available supply of sound-film subjects during the past few months. 

At the beginning of 1934, the available supply of 16-mm. sound- 
film entertainment subjects was perhaps less than 50 reels. At the 
present time at least ten times that number are actually circulating 
in the various libraries of the country. Sound-film service is avail- 
able from coast to coast, and it is no exaggeration to say that several 
thousands of reels of additional subjects are now available for repro- 
duction in 16-mm. size and will be in circulation just as quickly as 
the distribution of equipment arouses even a moderate demand for 
such a supply. 

Most of the 16-mm. sound-film projection equipment thus far sold 
has been for industrial purposes ; that is, for use by prominent manu- 
facturers in showing their own commercial pictures, made to carry 
publicity or make sales for the products featured. Recently a de- 
cided interest in equipment has developed among institutions that 
desire to use sound-films but could not do so until the fire hazard and 
expense of 35-mm. prints and projectors had been replaced by the 
safety, simplicity, economy, and portability of their 16-mm. succes- 
sors. 

In the days of the silent picture, non-theatrical exhibitions lacked 
the charm and emotional appeal furnished by the orchestral accom- 



178 W. B. COOK [j. s. M. p. E. 

paniment so cleverly cued to the picture in all the best theaters; 
but with the advent of 16-mm. sound-film, the family in the home, 
the children in the church, their parents at the club, or the shut-ins 
in the hospitals or institutions could for the first time enjoy every 
illusion of reality, previously enjoyed only by the spectator in the 
theater. The emotional appeal of combined sound and speech was 
beyond description and for the first time there was available a means 
of entertainment challenging the radio in its universal applications. 

What of the future of 16-mm. sound-film? As a matter of fact, 
its future has already been firmly assured. Its present status is far 
ahead of that of the 35-mm. at a corresponding stage of development. 
Whereas the 35-mm. sound-film had to start from nothing, and an 
entirely new art, science, and manufactured product had to be de- 
veloped, 16-mm. sound has but to follow closely in the footsteps al- 
ready carved by its older and bigger brother. It should never seek 
to compete in the professional amusement field, but will doubtless 
completely encompass the entire non- theatrical field. 

As any 35-mm. negative is capable of making 16-mm. reduction 
prints of quality comparable with their size, it is evident that any 
great film epic can be made available if the demand should justify it. 
Many of the greatest silent pictures were reproduced on 16-mm. film, 
and it is already evident that the 16-mm. sound-film will enjoy a 
popularity never imagined in the silent field. 

But future progress will be dependent upon the equipment manu- 
facturers. The film reproduction problems are solved. Projectors 
must be perfected to such a state that no skill is required and adjust- 
ments shall seldom be necessary; made so simple that any member of 
the family or any school-boy can operate them successfully; so silent 
and unobtrusive as to become a piece of household furniture, like the 
phonograph or radio. The present frequency response must be 
broadened so as to achieve sound reproduction on a par, at least, with 
that now attained in the average theater. And, perhaps most im- 
portant of all, ultimate success will depend largely upon the cost of 
the equipment and its film service. 

DISCUSSION 

MR. CRABTREE: Assuming a given picture, on 35- and 16-mm. film, would 
people rather go to the theater, or stay at home to see it? It seems to me that 
people more than ever these days will not stay at home. What is your experience ? 

MR. COOK: I can not answer that definitely. It obviously depends upon the 
disposition and temperament of the individual family. I am inclined to believe 



Feb., 1935] 16-MM. SOUND-FILM OUTLOOK 179 

that the American people prefer to leave their homes and flock into places of 
amusement. That is not so in Great Britain, where the home libraries have at- 
tained a volume of business so greatly surpassing what we find in this country 
that we are forced to conclude that national characteristics must have much to 
do with it. 

MR. LANE : What is the demand for color in 16-mm. sound-films? 

MR. COOK: So far there has been no demand for sound combined with color. 
Our library experimented with color in silent pictures a year or more ago, and 
found that the demand for color in 16-mm. silent pictures was extremely limited. 
The demand was not encouraged by the fact that the color that was attainable 
at that time was distinctly inferior, but I believe that the interest shown in in- 
quiring for the prints before they knew of their quality was rather limited. 

MR. PALMER: Do you believe that a sufficient demand might be created for 
16-mm. amusement subjects for the home to warrant producing subjects for that 
field? 

MR. COOK: So far as regards the home exclusively, that is open to question. 
But if other non-theatrical activities were included, institutional and the like, 
I should say that it was a prospective field. In fact, it is already expanding at a 
rate that surprises some of our optimistic proponents of the field. The interest 
is certainly great, and those who see and hear the available productions, as a 
rule, are enthusiastic in their reactions to them. 

MR. ROSENBERGER: What proportion of the demand is for theatrical, purely 
entertainment pictures, and for educational pictures, including semi-scientific, 
geographical films, biological films, and the like? 

MR. COOK: So far as the experience of a library founded principally for 
supplying entertainment is concerned, I can say only that our demand for educa- 
tional films is almost nil; and for scientific, even less. But it should be remem- 
bered that the libraries that have been established have appealed almost entirely 
to the urge for entertainment. It is rather difficult to know whether an ex- 
clusively scientific or educational library would receive a better response than we 
have. I believe it can be fairly accepted that the schools seem to have a pref- 
erence for entertainment films. 

MR. GOLDEN: I do not quite agree that the demand for educational films is 
practically nil, because my office has received thousands of letters from schools 
asking for sources of purely educational films. We maintain a bulletin service 
that is distributed on a subscription basis, listing the names of films of an industrial 
character. We also attempt to get as many of the educational films as are avail- 
able. I do know from conversations with Dr. Koon, of the Bureau of Education, 
U. S. Department of the Interior, that they are very much interested in the strictly 
educational film for classroom and curriculum use. The main difficulty, accord- 
ing to the Department of the Interior, is that they claim they can not get sufficient 
educational films, so as to be able to recommend them to the school systems of the 
country, which in turn hampers the sale of the 16-millimeter sound-film projectors 
that have been developed in the past two years. I believe that the Bureau of 
Education is about at the stage at which, if they can be shown that sufficient 
educational films are available for use in the schools, they can benefit the manu- 
facturers of 16-millimeter sound equipment considerably by endorsing the use of 
motion pictures in the classroom. 



A PHYSICAL DENSITOMETER FOR SOUND PROCESSING 
LABORATORIES* 

F. L. EICH** 



Summary. A simple densitometer for use in sound processing laboratories is 
described, employing the Weston photronic cell and a 50-cp., 12-v. lamp. The cell cur- 
rent is measured by a microammeter, suitably damped. The instrument retains its 
calibration well, densities can be measured within 0.005, and readings are consistently 
duplicated. 

The applications of light-sensitive devices in industry have been 
manifold. The development of an efficient photoelectric cell which 
does not require any external energy supply besides the light incident 
upon it has increased the number of uses for photoelectric cells. 
Although prior to the introduction of this type of cell upon the 
market, it was possible to make a physical densitometer using the 
other types of cells, nevertheless it always has entailed an additional 
amount of equipment, which introduced too many variables and made 
it unsuitable for the commercial field. 

The densitometer herein described has simplicity to recommend it 
without any sacrifice of accuracy. The device has been in constant 
service for a year at Paramount Laboratory and during that time 
has held to standard without adjustment (Figs. 1 and 2). 

A Weston photronic cell is used as the photosensitive device. 
The light source consists of a 50-cp., 12- volt lamp mounted in a re- 
flector and operated by a storage-battery. A condensing lens throws 
the light beam onto a plate with a rectangular opening 0.090 by Vie 
inch, and an opal glass is placed beneath the opening to diffuse the 
light. A rectangular aperture is preferable because it integrates 
the density area of the sound-track better than a circular opening. 
Over this aperture the density to be measured is placed. The 
photoelectric cell is mounted upon a hinged arm which raises the cell 
to allow the placing of the film over the light aperture. A shield is 
placed around the cell so that when it is in the reading position no 

* Received August 1, 1934. 

** Paramount Productions Inc.. Hollywood, Calif 
180 



A PHYSICAL DENSITOMETER 



181 



light but that passing through the film reaches the cell. The photo- 
electric cell current is registered by a Weston model 440 microammeter 
with a 30-microampere range. A suitable shunt is placed across the 
meter to provide the desired damping. The characteristic of the cell 
is such that greatest efficiency is attained when using a low load re- 
sistance. For this reason a model 440 is preferable. The scale of 
the meter is calibrated in density from to 1.0, which is sufficient for 
reading sound-track densities and sound negative gammas of the 
variable -density type of processing. 




FIG. 1. The densitometer fitted to the examining 
bench. 



In operation, the light falling upon the cell in the reading position 
is adjusted to register zero density on the meter without any film over 
the aperture. After this adjustment all that is required is to place 
the film over the aperture and read the density indicated by the meter. 

The device is calibrated in two ways. With low load resistance, the 
cell current increases linearly per unit of light intensity over a large 
range of intensity Therefore, using a diffused light source, the 
relation between current and diffused film transmission will also be 
linear (Fig. 3). Hence by measuring known densities with the cell 
and calculating the densities from the current vs. transmission curve, 
a calibration of the device is obtained. A secondary check is made by 



182 



F. L. EICH 



[J. S. M. P. E. 



measuring densities on the Bausch & Lomb polarizing densitometer 
and then by the photoelectric device (Fig. 4). 

The accuracy of the device has been found to be within 0.005 in 
density, and readings are consistently duplicated. Ease of opera- 
tion and the increased accuracy over the visual type of densitometer 




FIG. 2. Details of densitometer: (1) photronic cell; 
(2) aperture; (3) condensing lens; (4) reflector and lamp; 
(5) microammeter; (6) current supply for lamp. 

are great advantages. Paramount, of course, is the fact that visual 
fatigue, which is an important factor in measuring densities with the 
visual type of densitometer, is eliminated. A disadvantage of the 
single-range instrument is that densities above 1.0 fall within a very 
narrow portion of the meter scale, and measurements of those densi- 
ties are therefore not very accurate. However, by using a more sensi- 
tive meter and constant-resistance shunt networks of the attenuator 
type used in communication work, multiple ranges can be attained. 



Feb., 1935] A PHYSICAL DENSITOMETER 

too 



183 



<40 



10 



4 8 It 16 20 24 28 BZ 
MICROAMPERES 

FIG. 3. Calibration curve of densitometer. 



1.0 



u 

P 



.8 



u 

O .6 

J 

8 .4 



.2 4 .6 .8 l.O 

B.&L. DENSITIES (DIFFUSE} 

FIG. 4. Secondary check of densitometer. 



SOCIETY ANNOUNCEMENTS 
BOARD OF GOVERNORS 

At a meeting held on January 11 at New York, N. Y., further plans for the 
Hollywood Convention in May, as described below, were evolved. The fiscal 
report of the Society for 1934 was presented to the Board by Mr. O. M. Glunt, 
Financial Vice-President, indicating a satisfactory trend in the Society's affairs 
and the ability to improve its service to the membership and the industry. The 
JOURNAL was enlarged by fifty per cent for 1935, and generous financial assistance 
was provided for the Local Sections to enlarge their activities and hold more at- 
tractive and interesting monthly meetings. In addition, appropriations were 
made for the JOURNAL Award and the Progress Medal, the design of which latter, 
submitted by Mr. J. I. Crab tree, Editorial Vice-President, was approved by the 
Board. Other actions taken by the Board, as regards the Sectional Committee 
on Motion Pictures, under the A. S. A., lapel buttons, etc., were as described below. 

HOLLYWOOD CONVENTION 

As announced previously, the Spring Convention will be held this year at 
Hollywood, May 20-24, inclusive, headquarters at the Hotel Roosevelt. Mem- 
bers of the Society are urged to make every effort to attend, and contribute to 
making this convention the greatest and most interesting in the history of the 
Society. The Pacific Coast Section Board of Managers, under the chairmanship 
of Mr. G. F. Rackett, and with the Assistance of Messrs. E. Huse, Executive 
Vice-President of the Society, and W. C. Kunzmann, Convention Vice-President, 
are collaborating in arranging the details of the Convention and the Apparatus 
Exhibit. 

The latter will be known as the "Studio Practice and Equipment Exhibit" and it 
is hoped that the contributions of the studios, in displaying the developments and 
advances in technic and equipment of the studios, will be quite extensive. Plans 
are being made for an interesting program of technical papers, and Messrs. J. I. 
Crabtree and J. O. Baker, Chairman of the Papers Committee, promise a number 
of outstanding demonstrations and presentations. Several technical sessions will 
be held in the evenings, in order to permit those to attend whose employment 
would not permit them to attend the day sessions. Visits to several of the 
studios are being arranged for the open afternoons. 

ATLANTIC COAST SECTION 

The first monthly meeting of the Section was held in the auditorium of the 
Electrical Association of New York on January 9, at which Mr. Rudolph Wolf, of 
Electrical Research Products, Inc., presented a paper entitled "Visual Accom- 
paniment." The paper presented new ideas concerning the production of pic- 
tures to accompany music, rather than the usual method of arranging the music to 
supplement the picture. 
184 



SOCIETY ANNOUNCEMENTS 185 

As demonstrations of the principles involved, there were first projected two 
films produced by the Savage method, viz., The Unfinished Symphony and Les 
Preludes; followed by three examples of "Musical Moods," viz., Fingal's Cave, 
Italian Caprice, and Barcarole. All the films were done in Technicolor. 

The meeting was well attended, and the presentation aroused considerable in- 
terest and discussion. 

SECTION COMMITTEE ON MOTION PICTURES UNDER THE A. S. A. 

As reported previously, a Sectional Committee on Motion Pictures, organized 
according to the procedure of the American Standards Association, is being es- 
tablished, with the Society of Motion Picture Engineers as sponsor. At the 
meeting of the Board of Governors on January 11, a list of those organizations, 
firms, and societies recommended for representation upon the Sectional Com- 
mittee was approved, and invitations are now being mailed to those bodies to 
name their representatives to the Committee. As soon as the Committee will 
have been formed, an organization meeting will be called, a chairman appointed or 
elected, and an agenda of projects requiring study for possible and needful stand- 
ardization will be drafted. 

[ SOCIETY SUPPLIES 

Certificates of Membership may be obtained from the General Office by all 
members for the price of one dollar. Lapel buttons of the Society's insignia are 
also available at the same price. 

Black fabrikoid binders, lettered in gold, designed to hold a year's supply of the 
JOURNAL, may be obtained from the General office for two dollars each. The 
purchaser's name and the volume number may be lettered in gold upon the back- 
bone of the binder at an additional charge of fifty cents each. 

Requests for any of these supplies should be directed to the General Office of 
the Society at the Hotel Pennsylvania, New York, N. Y., accompanied by the 
appropriate remittance. 



STANDARD S. M. P. E. 

VISUAL AND SOUND TEST REELS 

Prepared under the Supervision 

OF THE 
PROJECTION PRACTICE COMMITTEE 

OF THE 
SOCIETY OF MOTION PICTURE ENGINEERS 



Two reels, each approximately 500 feet long, of specially pre- 
pared film, designed to be used as a precision instrument in 
theaters, review rooms, exchanges, laboratories, and the like 
for testing the performance of projectors. The visual section 
includes special targets with the aid of which travel-ghost, 
lens aberration, definition, and film weave may be detected 
and corrected. The sound section includes recordings of 
various kinds of music and voice, in addition to constant 
frequency, constant amplitude recordings which may be used 
for testing the quality of reproduction, the frequency range 
of the reproducer, the presence of flutter and 60-cycle or 96- 
cycle modulation, and the adjustment of the sound track. 
Reels sold complete only (no short sections). 

PRICE $37.50 FOR EACH SECTION, 
INCLUDING INSTRUCTIONS 

(Shipped to any point in the United States) 

Address the 

SOCIETY OF MOTION PICTURE ENGINEERS 

HOTEL PENNSYLVANIA 
NEW YORK, N. Y. 



JOURNAL 



OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 

Volume XXIV MARCH, 1935 Number 3 



CONTENTS 

Page 

Light Source Requirements for Picture Projection 

F. E. CARLSON 189 

A Mechanical Demonstration of the Properties of Wave Filters . 

C. E. LANE 206 

Electronic Tube Control for Theater Lighting 

J. R. MANHEIMER AND T. H. JOSEPH 221 

Roentgen Cinematography R. F. JAMES 233 

My Part in the Development of the Motion Picture Projector . . 

T. ARMAT 241 

Overcoming Limitations to Learning with the Sound Motion 
Picture V. C. ARNSPIGER 257 

A Roller Developing Rack for Continuously Moving the Film 
during Processing by the Rack-and-Tank System . . C. E. IVES 261 

William Van Doren Kelley 275 

Spring, 1935, Convention at Hollywood, Calif 278 

Society Announcements 281 



JOURNAL 

OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 



SYLVAN HARRIS, EDITOR 

Board of Editors 
J. I. CRABTREE, Chairman 

O. M. GLUNT A. C. HARDY L. A. JONES 

J. O. BAKER 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1935, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 



Officers of the Society 

President: HOMER G. TASKBR, 4139 38th St., Long Island City, N. Y. 
Past-President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
Executive Vice-President: EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, 

Calif. 

Engineering Vice-President: LOYD A. JONES, Kodak Park, Rochester, N. Y. 
Editorial Vice-P resident: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Financial Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Convention Vice-President: WILLIAM C. KUNZMANN, Box 6087, Cleveland, Ohio. 
Secretary: JOHN H. KURLANDER, 2 Clear-field Ave., Boomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York, N. Y. 

Governors 

MAX C. BATSEL, Front & Market Sts., Camden, N. J. 
LAWRENCE W. DAVEE, 250 W. 57th St., New York, N. Y. 
ARTHUR S. DICKINSON, 28 W. 44th St., New York, N. Y. 
HERBERT GRIFFIN, 90 Gold St., New York, N. Y. 
WILBUR B. RAYTON, 635 St. Paul St., Rochester, N. Y. 
SIDNEY K. WOLF, 250 W. 57th St., New York, N. Y. 



LIGHT SOURCE REQUIREMENTS FOR PICTURE 
PROJECTION* 

F. E. CARLSON** 

Summary. The light sources available for motion picture projection are discussed 
in relation to their performance characteristics and their application to existing equip- 
ment and conditions of service. The optical systems used with the sources, the usable 
size and shape of the source, the brightness and light output of the filament, and ven- 
tilation are discussed. 

During the past few years, important improvements in lamps and 
equipment have extended the scope of application of the several types 
of motion picture projectors. These improvements have been so 
far-reaching in their effect that a restatement of the factors deter- 
mining the lighting performance of projectors may be of interest and 
value. 

Desired size and brightness of projected picture, and the dimensions 
of the aperture or film through which light must be directed, form the 
starting point from which all development of optical systems and light 
sources must proceed. The designer is interested in the optimum 
result that can be achieved in each case. But he must also give 
consideration to matters of cost, of size and weight, and to operating 
features acceptable in the different classes of service. To form the 
best judgment of what would constitute a successful device for a 
given market, he must make as exact a determination as possible of 
the gains and losses involved in modifications undertaken because of 
economic and convenience considerations. 

In Table I the more common markets or services are listed in con- 
junction with the three sizes of film that are applicable. The mini- 
mum light required in each case is indicated, together with the prac- 
tical limitations as to the bulb size and operating features that ex- 
perience has shown to be desirable. The most desirable values of 
light output are not so well established, but maximum acceptable 
values are in every case several times the minimum figures given. 

* Presented at the Spring, 1934, Meeting at Atlantic City, N. J. 
** Incandescent Lamp Dept., General Electric Co., Cleveland, Ohio. 

189 



190 



F. E. CARLSON 



OPTICAL SYSTEMS 



[J. S. M. P. E. 



Some consideration of the problems involved in applying lens 
combinations to picture projection is essential to an understanding of 
light source requirements. The focal length and limiting relative 
apertures (//value) of the projection lens are fairly well established 
for each film (aperture) size employed. Increasing the relative aper- 
ture of an objective lens makes possible material improvement in 
screen illumination, because the light passed is inversely proportional 
to the square of the //value of the lens. In the case of the 35-mm. 
film equipments, a value of //2.0 is usually the limit. 

TABLE I 

Classification of Projection Service 



Film Size and 
Type of Service 


Picture 
Width, 
Range- 
Feet 


Minimum 
Acceptable 
Screen 
Lumens 

(No 
Shutter) 


Maximum 
Limit of 
Bulb Size 


Desirable 
Lamp 
Voltage 


Approx. 
Equip. 
Cost Not 
Including 
Sound $ 


8-mm. 












Toy 


1-2 


5 


T-8 or S-ll 


100-120-Volt 


10-iO 


Home 


1-3 


20 


T-8 or T-10 


Lamps Pre- 


25-100 


16-mm. 








ferred but 




Home 


3-6 


50 


T-8 or T-10 


Low-Voltage 


25-200 










Lamps Feasible 




Educational and 












Business 


4-8 


150 


T-10 or T-12 


100-120 


100-200 


Small Auditorium 


6-10 


250 


T-12 


Any 


100-300 


35-mm. 












Portable E duca- 












tional and Busi- 












ness 


8-12 


300 


Short T-20 


100-120 


100-375 


Semi - Portable 












Auditoriums 


10-12 


450 


Long T-20 


100-120 


200-500 


Theater 


10-20 


800 


T-24 


Any 


500- 



In the case of 8- and 16-mm. film objectives, the//2.0 size has been 
most generally used. However, it has been found that for these 
services relative aperture values up to //1. 6 can be successfully em- 
ployed. 

If the projection lens is to pass the greatest amount of light, the 
source image must fill the entrance pupil of the lens. A uniformly 
lighted screen is also desired. It follows that for a given aperture and 
projection lens, the minimum size of condenser and the relation be- 
tween the condenser diameter and the condenser-aperture spacing 
must be such that equal areas of condenser are visible through the 



Mar., 1935] 



LIGHT SOURCES FOR PROJECTION 



191 



optical system from all points of the screen. In the schematic dia- 
gram of Fig. 1, lines AI and A '/' define this relation. 

Assuming equal pick-up angle, the smaller the condenser, the 
smaller are the dimensions of the source from which the light can be 
redirected into a beam through the aperture to the projection lens. 
On the other hand, the losses at the aperture are minimized as the 
size of the condenser is reduced. Thus the particular combination of 
condenser diameter and condenser-aperture spacing employed must 
be a compromise between efficiency of utilization and the size of the 
source required for supplying the desired amount of light. The 
available light from a given source is utilized most efficiently with a 
condenser of maximum refracting power; that is, with the shortest 




FIG. 1. Relation between condenser diameter and condenser- 
aperture spacing for uniformly lighted screen. 

source-condenser spacing. Higher levels of screen illumination are, 
however, often attainable with sources of higher wattage and light 
output at greater distance. 

These various considerations, having a definite bearing upon the 
wattage of the source, the type of the optical system, and the size and 
weight of the projector, have resulted in the typical systems upon 
which the following data are based. 

USABLE SOURCE SIZE AND SHAPE 

As pointed out above, the size of the source that the optical system 
can utilize depends upon the magnification of the system and the 
entrance pupil of the projection lens. The shape of the utilized source 
depends upon the position of the source image; it is round if it falls 



192 



F. E. CARLSON 



[J. S. M. P. E. 



within the projection lens, and it approaches the rectangular shape 
of the aperture if it falls between the lens and the aperture. 

Back-testing provides a simple method of determining the size and 
shape of the source that an actual optical system can utilize. This is 
done by placing a large diffusing source of light in front of the projec- 
tion lens and a target for holding photographically sensitized paper at 
the source position. The diffusing source produces a spot of light 
upon the target that conforms in shape and dimensions to the area of 
the source that is effective. Table II presents a summary of such an 
analysis of typical optical systems as used in the several types of 
projectors. Data on stereopticon projection systems are not included 
because, within reasonable limits, source dimensions are not a govern- 
ing factor in that service. 

TABLE II 

Summary of Back-Test on Typical Optical Systems 



Type of Optical System 


Av. Source 
Condenser 
Spacing 
(Mm.) 


Projection Lens 
E. F. //Value 


Shape 


Back-Test 
Av. Dimen- 
sions (Mm.) 


8 -mm. 


15 


r 


//2.0 


Oval 


6.0 




18 


1" 


//2.0 


Oval 


7.0 


16-mm. 












Spherical Condenser 


18 


2" 


//2.0 


Round 


10.0 


Aspheric Condenser 


18 


2" 


//2.0 


Round 


to Oval 7 . 5 


Aspheric Condenser 


18 


2" 


//I. 65 


Round 


to Oval 8.5 


Aspheric Condenser 


21 


2" 


//I. 65 


Round 


to Oval 10.0 


35-mm 












Prismatic 


40 


5" 


Series 1 


Oblong 


12 X 7.5 


Spheric 


40 


4" 


Series 1 


Round 


15-20 


Aspheric 


40 


5 3 // 


Series 2 


Oblong 


11-14 



In addition to their importance from the standpoint of determining 
requisite source dimensions, these data are of interest as confirmation 
of some of the general statements made above. It is to be noted 
that, for the same relative aperture of projection lens, aspheric sys- 
tems do not utilize as large a source area as do spherical condenser 
systems; their high efficiency is due to the large solid angle from 
which they are able to refract the light satisfactorily. On the other 
hand, the use of projection lenses of larger relative aperture increases 
the dimensions of the usable source with either type of condensing 
system. 



Mar., 1935] LIGHT SOURCES FOR PROJECTION 193 

BRIGHTNESS OF FILAMENT SOURCES 

The source problem is thus one of providing the greatest total light 
output from the utilizable space consistent with feasible bulb size and 
acceptable lamp cost and performance 

Brightness of source is dependent upon two things, the operating 
temperature of the filament and the amount of incandescent fila- 
ment that can be disposed in a unit area. Temperature of the fila- 
ment, and hence light output and brightness, fortunately increase 
rapidly as the source is designed for shorter life. For a given lamp 
life, filament temperature may be increased by the use of larger 
(thicker) filament wire. Thicker filament means either increased 
wattage, or reduced voltage and higher current. There is a limit 
beyond which low voltage and high current become disadvantageous 
due to the heavy heat conduction losses through the larger lead-in 
wires required. 

The amount of incandescent filament that can be disposed within a 
unit area depends, aside from diameter, upon the filament construc- 
tion employed. Filaments can be concentrated by coiling; in fact, a 
coil may be coiled upon a second mandril to form a coiled-coil. The 
coils can be disposed in more than one plane, to fill the open spaces 
that would otherwise be presented to the lens. Such constructions 
have become practicable due to the progress made in recent years in 
filament heat-treating processes. More favorable methods of 
mounting have also contributed to preventing distortion from the 
great stresses of heating and cooling. Thus these more efficiently 
disposed filaments are kept in place throughout the life of the lamp. 

The concentration of filament is attained with some loss in effi- 
ciency of light generation because the temperature at various points 
of the filament is thereby made less uniform. It is the hottest point 
that determines lamp life, whereas the total light output depends 
upon the average temperature. The ratio of maximum to average 
temperature is greater for a coiled-coil filament than for a single 
coil, and still higher for a biplane source, which thus has the lowest 
efficiency in light generation although not in utilization. 

The use of a spherical mirror behind the filament is important in 
achieving a more uniformly bright source and hence higher average 
source brightness. The increase is obviously not the same for all 
filament constructions, since they vary as to the open spaces to be 
filled by the reflected coil images. Thus the mirror increases the 
effective brightness of monoplane sources of the coiled-coil type by 



194 



F. E. CARLSON 



[J. S. M. P. E. 



upward of 45 per cent, of monoplane sources of the coiled-coil type 
by about 35 per cent, and of biplane sources not more than 30 per 
cent. 

Many of the heat and light rays reflected from the mirror are inter- 
cepted by the filament and increase its temperature. On the other 
hand, this means a slightly higher light output for a given wattage. 
Unfortunately, it also increases the disparity between maximum and 
average temperatures. Both these factors must be taken into ac- 
count in the design of the lamp. 

TABLE III 

Light Output per Unit of Source Area 



Source Construction 


Watts 


Volts 


Average Source 
Dimensions (Mm.) 


Life 


Cp. 

per 
Sq. Mm.* 


Monoplane (CC13) 


200 


100-120 


7.8 X 8.1 


50 


10.2 


Monoplane (2-Seg-CC8) 


200 


100-120 


5.9 X 6.0 


25 


16.5 


Monoplane (CIS) 


500 


100-120 


13.0 X 12.0 


50 


13.5 


Monoplane (13} 


500 


100-120 


10.4 X 10.0 


25 


20.2 


Monoplane (CIS) 


500 


30 


9.8X 9.2 


50 


23.9 


Monoplane (CIS) 


500 


30 


8.6 X 8.3 


25 


31.2 


Monoplane (C13) 


900 


30 


10.7 X 13.8 


100 


27.6 


Biplane (C13D) 


500 


100-120 


8.2 X 8.2 


25 


30.8 


Biplane (C13D) 


750 


100-120 


9.3 X 9.9 


25 


34.9 


Biplane (C13D) 


1000 


100-120 


11.0 X 12.1 


25 


35.2 



The net effect of all factors upon the brightness is indicated by the 
data of Table III for the source forms that are in general most ad- 
vantageous. It will be observed that the advantage of low voltage is 
hardly sufficient to outweigh the disadvantage of the weight and 
cost of auxiliary equipment in the classes of service in which portable 
projectors are usually employed. As to lamp life, the advantages of 
higher screen brightness outweigh the replacement cost for portable 
projectors used intermittently or for occasional service, whereas for 
continuous service in large projectors, as in the theater, replace- 
ment cost is a major factor and therefore longer lamp lives prevail. 

Those portions of the source near the optical axis are more ef- 
ficiently utilized than those farther from the axis. The lower portion 
of Fig. 2 shows the results of tests with sources of various sizes; the 
percentage of generated lumens reaching the screen (optical efficiency) 

* These values apply when a spherical mirror is used and are an average 
through an angle of 100 degrees. 



Mar., 1935] 



LIGHT SOURCES FOR PROJECTION 



195 



is plotted against the source dimensions. Although the data are 
those for a typical 16-mm. spherical condenser system with//2 pro- 
jection lens, the same general relations hold for other optical systems. 
It will be noted that the data apply for a system that back-tested 
to a circular source of 9.8 mm. Yet the screen-lumen curve continues 
to rise for sources of greater diameter, illustrating the fact that the 
margins of the source are cooler than the center and that therefore a 
source of somewhat greater dimensions actually adds more light to the 
screen by bringing up the average brightness of the part used. 




!i 



4 6 10 

Average Source Dimmion (MM) 



FIG. 2. Effect of optical efficiency on initial screen illumination 
with monoplane and biplane sources (typical 16-mm. spherical con- 
densing system with//2.0 projection lens) . 

PROJECTION LAMPS 

Up to this point optical systems and source characteristics have 
been considered without regard to wattage. In practical applica- 
tion of these data in the selection of a lamp best suited to a given 
service other factors such as bulb limitations and cost must also be 
considered. 

If the wattage is plotted against the source size, as in Fig. 3, it will 
be noted that the source producing substantially maximum screen 
illumination in that type of equipment consumes 750 watts. This 
particular optical system is designed for a source-condenser spacing 



196 



F. E. CARLSON 



[J. S. M. P.E. 



that necessitates the use of either a centered-filament T-10 bulb or a 
bulb of larger size with offset filament. This introduces problems of 
bulb temperatures and bulb blackening. 

The limiting wattage for each bulb size depends upon the tem- 
perature at which the bulbs devitrify or soften and blister. This 
temperature varies with the glass employed. Both the volume of the 
bulb and the distance of the filament from the nearest wall govern the 
glass temperature, but the latter is the larger factor. Therefore, an 
offset filament in a larger bulb permits little gain in wattage, and 




2 4 6 10 12 14 

Average Source Dimension (MM) 

FIG. 3. Relation of initial screen illumination to wattage for 
monoplane and biplane sources (typical 16-mm. spherical condensing 
system with//2.0 projection lens). 

centered filament lamps in bulbs that will withstand high tempera- 
tures are to be preferred. 

Bulb volume is important from the standpoint of concentration of 
blackening and increase of glass temperature due to absorbed radia- 
tion. An increase in bulb temperature through life of as much as 30 
per cent occurs in extreme instances. Ventilation can do much to 
reduce bulb temperature, but there is a limit to what it can do be- 
cause in extreme cases concentration of blackening opposite the fila- 
ment occurs regardless of air volume. 

The adequacy of the average light output throughout the life of 
the lamp is of greater interest and importance than initial screen 




Mar., 1935] 



LIGHT SOURCES FOR PROJECTION 



197 



lumens in evaluating projection lamps. Their rate of blackening 
must also be taken into account. This is illustrated in Fig. 4. In a 
given bulb, a wattage lower than that giving the highest initial screen 
illumination will often give a better average value. 

Analyses for the various services similar to that represented by 
Fig. 4 have led to the following practice where sufficient ventilation 
is provided: 

Bulb Size Maximum Wattage 

*7%S 200 

T-10 500 

T-12 750 

T-20 (Short) 1000 

T-20 (Long) 1500 



T-24 



2500 



5 n 



g. 

,. 




^ 



RELATIVE SCREEN ILLUMINATION 
TYPICAL I6MM OPTICAL SYSTEMS 
VARIOUS PROJECTION LAMPS 



PROJECTION LAMP 

FIG. 4. Relative screen illumination. 

The glasses employed in lamps and the ventilating systems pro- 
vided in projectors of a few years ago would not have permitted a 
bulb of acceptable size for the total wattage that can now be con- 
centrated in the space effective in contributing light to the screen. It 
has now also become possible to utilize Pyrex bulbs, which can with- 
stand intermittent temperatures up to 550 to 600F. 



* T designates tubular bulbs; the numbers represent diameters in eighths of 
an inch. 



198 



F. E. CARLSON 



[J. S. M. p. E. 



VENTILATION 

More effective flow of a given volume of air has been the third 
important feature in making possible the much-improved screen 
results from the newer projectors. Some types of equipment were 
already providing large volumes of air to cool the lamps and further 
increases would have produced severe problems of noise, excessive 

blower speeds, or even bulky 
blowers. Smoke pictures, disclos- 
ing the distribution and direction 
of air flow through ducts and 
lamp houses, and measurements 
of their relation to bulb tempera- 
ture, furnish the key to efficient 
utilization of the air. The sys- 
tern illustrated in Fig. 5 provides 
adequate cooling for the new 
higher wattages, and accom- 
plishes it without larger blowers or 
higher speeds. Only the air close 
to the bulb is really effective, and 
since the bulb is hottest in the re- 
gion of the optical axis, restriction 
of the duct in this zone puts all 
the air to work and moves the air 
fastest where the bulb is hottest. 



TODAY'S PROJECTION LAMPS 

The projection lamps presently 
available represent in many re- 
spects the most advanced product 
of the highly developed incandes- 
cent lamp art. These amazingly 
compact units were designed in 
accordance with the above analysis to meet existing as well as ex- 
panding requirements of the several classes of service in both optical 
characteristics and cost. There are, in the lower wattages, mono- 
plane sources of single-coil filaments, highest in efficiency of light 
generation and lowest in cost; others in coiled-coil construction of 
greater concentration and slightly higher cost; and, finally, in the 
higher wattages, both monoplane and biplane sources giving every 




FIG. 5. Schematic diagram of sug- 
gested ventilation system for projection 
lamps. 



Mar., 1935] LlGHT SOURCES FOR PROJECTION 199 

class of projector user the most economical service for his particular 
requirements. These modern lamps have made it possible, except 
in the very largest units, to dispense with the cost and inconvenience 
of low-voltage operation for new equipments and to substitute 
standard- voltage lamps in the older projectors as well. 

DISCUSSION 

MR. PALMER: Is the advantage of air cooling principally to prevent blackening 
of the bulb, or to give a longer filament life? 

MR. CARLSON: Air cooling seems to be an unimportant factor in both bulb 
blackening and filament life, but forced ventilation is essential where high watt- 
ages are employed in a small bulb, if the glass is not to fail. Lamps that must 
operate in projectors having only natural ventilation are necessarily either of 
lower wattage for a particular bulb size or, for a given wattage, of larger diameter. 

MR. MITCHELL: The problem of the amateur is different from that of the 
theater. First, there is no question of box-office benefit involved. Then, the 
amateur requires a maximum of light source and a minimum of bulk. The effect 
of air cooling is three-fold: first, it decreases lamp blackening; second, it in- 
creases lamp life; third, and perhaps most important, it keeps the projector cool, 
which is definitely desirable in equipment used by amateurs. 

The amateur, and more particularly the industrial user, has pressed the manu- 
facturer for more light extra light is desired without appreciable increase in 
bulk. The researches referred to by Mr. Carlson make it possible to give the 
maximum effective light in the most compact form, and users of the small 16-mm. 
projectors have been able to show pictures of theatrical quality to fairly large 
audiences. For instance, with a 750-watt lamp we quite regularly show a 12- or 
14-ft. picture to 1000 or 2000 people with quite satisfactory results. 

MR. GAGE: Some time ago I had occasion to study condenser design for 
these small projectors to see what could be done, and I made a certain assump- 
tion, namely, that the filament of the lamp was going to be in the center of either 
a tubular or globular bulb. That then seemed reasonable for what seemed to be 
a necessary size of filament for adequate illumination such as is desired by the 
amateurs. It led to a bulb of a certain diameter, and therefore the glass of the 
condenser had to be placed outside the bulb. That led to a certain condenser 
design. 

Those who make the projectors like to have very small condensers. Now the 
whole thing was completely revised from our standpoint by what the lamp de- 
signers did, pushing the filament of the lamp close to one side of the bulb. 
Then the projector manufacturers found that they could use those 750-watt 
lamps if they were ventilated. I do not see any reason why that can not be done 
with the professional-size projectors if it is found that the optics of the system 
will work with the filament brightness available. We are not now limited as to 
the position of the filament in the lamp if we are willing to use forced ventilation. 

MR. BEGGS: The factors that Mr. Carlson presented are in good agreement 
with our findings, and therefore may be taken as representative. One of the most 
important items encountered in recent years is the limitation of bulb glass tern- 



200 F. E. CARLSON [J. S. M. p. E. 

perature. The purpose of the ventilation is primarily to prevent the tempera- 
ture of the glass from rising to a point at which it bulges and the bulb fails. 

Several years ago, smaller bulbs with higher wattages were introduced. Today 
they are the most common sources in amateur projectors of the higher power. 
The 500-watt lamp, for example, in the T-10 bulb, looks exactly like lamps of 
smaller wattage that do not require air cooling. The lamp manufacturers wrap 
around each such lamp a cautionary label advising the operator to use forced 
draft ventilation. 

This matter deserves special attention, because in some instances operators 
have used bulbs requiring air cooling in projectors that do no provide for it. 

MR. PALMER: Has the outside temperature of the bulb, anything to do 
with the life of the filament or with the blackening of the bulb? 

MR. CARLSON: Surrounding temperature has not been a factor in filament 
performance. However, surrounding temperature (and, therefore, bulb tempera- 
ture) might have some influence on the distribution of the blackening. The 
heated gas in the bulb naturally circulates, carrying the bulk of the evaporated 
tungsten to the upper portions of the bulb, where it forms a black deposit. If 
the tops of the lamps were cooled to a greater differential than at present, the gas 
currents would move more rapidly and more of the blackening would, theoreti- 
cally, occur at the top and less in the optical zone. This idea has not been sub- 
jected to practical demonstration and therefore the actual improvement in 
average light output throughout the life of the lamp has not been determined. 

MR. FRITTS: In recent years there has been continuous competition between 
the manufacturers of amateur projectors to provide more and more wattage in 
the lamp house without particular reference to the results on the screen. This 
has led to poor technical practice and more or less deception of the customer. 

PRESIDENT GOLDSMITH: Do you suggest that the rating or the name-plate 
of the projector carry a statement of screen illumination or some proportionate 
equivalent, rather than a statement of lamp wattage? 

MR. CARLSON: Illumination per unit of screen surface per watt applied to 
the light source. 

MR. HARDY: It is impossible to tell from the wattage of a lamp how much 
illumination will result from it on the screen. Projectors might be rated in terms 
of the size of the screen that they will illuminate to some specified level. I be- 
lieve the public is not entirely to blame for the confusion that exists. The public 
has attempted to find some rating that would indicate the power of the projector. 
The only rating they have been able to find is the wattage of the lamp. If the 
practice had developed of recommending the screen size, the public might have 
been quite content with that. 

MR. MITCHELL: There are quite a few practical difficulties in the way of 
rating the projectors. In industrial work, for instance, we consistently encounter 
cases wherein the user is in doubt as to whether to use, say, a 100-volt, 400-watt 
lamp on a 110-volt circuit, or a 110-volt, 500-watt lamp. In other words, he 
studies the total illumination as compared with lamp life and lamp cost. Some 
prefer more screen illumination even though it cuts down the life of the lamp. 
One man will be willing to pay for lamps, say, at the rate of 20 to 30 cents an hour 
whereas another will want the maximum illumination he can get at the lowest 
possible price and would object to a lamp cost of, say, 2 or 3 cents an hour. Even 



Mar., 1935] LIGHT SOURCES FOR PROJECTION 201 

the cost factor is not a criterion. Some satisfactory empirical ratio of cost to 
screen illumination might be found, but to a large extent each individual has a 
different idea of what would be satisfactory. 

MR. FRITTS: The wattages and screen illumination have been increased 
beyond what is needed in the average home, except for Kodacolor projection, even 
to the point where the black-and-white mixtures are ruined by the excessive 
illumination. But there exists the industrial field, the advertising field, where 
these larger wattages can be used to advantage with larger audiences. 

The division between those two fields needs to be considered; the matter of 
filament temperature and lamp life is different in the two fields. 

MR. TASKER: The problem of rating the optical efficiency of a projector 
seems to be a little like the horse-power rating of an automobile. We accept 
the horse-power ratings without knowing or caring about the conditions under 
which they were measured, hoping only that the conditions are the same for all 
manufacturers so that the rating will give a clue to performance. Similarly, 
the public is anxious to know how much light may be expected from the pro- 
jector, the purchase of which is being considered. The only criterion that the 
manufacturer has as yet given the customer is that of lamp wattage, which 
utterly disregards the very important factor of optical efficiency. 

It should not be difficult for the Society to work out a so-calledS. M. P. E. rating 
for projectors which would indicate the relative screen brilliancy to be expected 
from a projector under a standard set of conditions as to lamp current, lamp volt- 
age, screen size, etc. The public need not have any understanding as to what 
the technical conditions may be so long as they have reasonable assurance that 
the rating so determined is a criterion of the optical performance which they may 
expect from the projector. The result of such a measure should be to turn the 
competitive efforts of the manufacturers into a race for more efficient rather than 
more wasteful use of the customer's projection lamp dollars. 

MR. FARNHAM: Unfortunately, the obtaining of greater screen intensity is 
not the simple matter of increasing the lamp wattage. As Mr. Carlson has shown, 
increasing the lamp wattage results hi more rapid blackening of the bulb, and if the 
projector ventilating system is inadequate, the bulb blisters. Also, for a par- 
ticular filament form the source area increases with increased wattage. If a 
special source construction were resorted to in order to minimize this, the cost 
of the lamp would go up. Five-dollar lamps can, for example, hardly be standard 
with fifteen- to twenty-dollar projectors. The factors of lamp wattage, bulb size, 
optical system design, ventilation, and lamp and projector cost are closely inter- 
related, and one can not be altered without affecting the others. 

We have gone through a period of rather intensive study and development in 
lamps, optics, and projectors, and it is not surprising that it has led to a some- 
what feverish commercial exploitation which has not always waited for the more 
complete answer. But by systematic and cooperative effort, we have arrived 
at a much better understanding of the limits of good practice, so that, although 
progress will continue, there is every indication that it will be more systematic, 
less rapid, and less upsetting than in the past few years. 

MR. PORTER: Unquestionably it would be advantageous to both the purchas- 
ing public and the projector manufacturers to have some system of rating pro- 
jectors that would show at a glance then- limitations. The problem is com- 



202 F. E. CARLSON (J. S. M. p. E. 

plicated, but no more so than an almost identical one in connection with the manu- 
facture and sale of sunlamps for health purposes. Such lamps used to be rated in 
accordance with the length of time required to produce a slight reddening of the 
skin (the beginning of sunburn, or erythema), with no account taken of the area 
covered by the irradiation. Consequently, the total energy received by the 
subject under treatment was directly proportional to the irradiated area. A lamp 
producing erythema on an area the size of a penny would have just as high a 
rating as one that would produce erythema over half one's entire body in the same 
period of time. 

That situation was no more ridiculous than rating projectors in accordance with 
the wattage of the lamp. A new rating was developed for sunlamps which ap- 
praised them in accordance with the total amount of biologically effective energy 
delivered on a definite area at a given distance from the lamp. The sunlamp 
manufacturers then agreed to classify the various lamps into 6 divisions, each 
division having a maximum and a minimum energy limit. A neutral testing 
laboratory was chosen to rate sunlamps for the various manufacturers. Each 
new type of lamp produced is rated by that laboratory and all lamps sold have 
tags attached to them showing the entire 6 divisions and stating into which class 
the particular lamp falls. 

The system has worked out very well with sunlamps, and I am sure a similar 
set-up could be developed for motion picture projectors. First it would be neces- 
sary to classify all makes and models of projectors as to their theoretical maximum 
possible output as limited by their optical systems. This could be done by measur- 
ing the total lumen output of the projector without film but with shutter running. 
For the purpose a uniform diffusive light source of fixed brilliancy (such as a 
piece of diffusing opal glass) entirely filling the aperture plate opening would be 
used. The degree of diffusion and brilliancy of this secondary test source 
would have to be standardized by the testing laboratory and agreed upon by the 
various manufacturers. 

Such tests would then establish the theoretical maximum possibilities of the 
various projectors. The next step would be to test the projectors with all the 
available standard projection lamps that might be used in each projector. Here 
again the testing laboratory would have to select representative lamps, setting 
limits as to source size, lumen output, etc., which limits should agree with the 
lamp manufacturers' ratings. Having made these tests the projectors could then 
be assigned percentage values representing the percentage of the theoretical 
maximum attained when the projectors were equipped with various styles of 
standard lamps. 

With these data available the projector manufacturers could agree to certain 
classifications of projectors with maximum and minimum limits of lumen output 
for each classification. They could publish this classification together with the 
percentage tables. The lamp manufacturers could supplement the data with 
information as to the effect upon life and lumen output of operating the various 
lamps' over-voltage, indicating such cases as might be likely to result in blistered 
bulbs if this were done, indicating also the increase in rate of lumen output de- 
preciation due to increased blackening. Data could also be given as to the foot- 
candle intensities on screens of various sizes that could be attained with any 
number of lumens. 



Mar., 1935] 



LIGHT SOURCES FOR PROJECTION 



203 



We would then have tables looking somewhat like the following ones. As 
changes were made in projectors or new ones developed or brought out, they 
could be added to the data from year to year. The figures given are not accurate, 
but merely illustrate what the tables might look like. 



Projector Type 


Mfgrs.' Nai 


Class 




8-mm. E 


A Co. 


16-mm. B 


X Co. 


16-mm. B 


PCo. 


16-mm. A 


PCo. 


etc. 


etc. 



Theoretical Max. 



200 

450 
500 
475 
etc. 



Per Cent of Theoretical Max. 
with Lamps 



A 

40 



B 

65 



etc. 



C D E F G 



60 



80 



etc. 



73 



etc. 



Lamp 


Watts 


Volts 


Approx. 
Amperes 


Bulb 


Base 


Initial 
Lumens 


Hours' 
Life 


A 


50 


115 


4.3 


T-8 


SCBay 


790 


50 


B 


100 


115 


0.87 


T-8 


SCBay 


1,870 


50 


C 


200 


115 


1.8 


T-10 


Med. P. F. 


4,000 


50 


D 


250 


50 


5.0 


T-10 


Med. P. F. 


5,950 


50 


E 


300 


115 


2.6 


T-10 


Med. P. F. 


7,200 


25 


F 


500 


115 




T-20 


Med. P. F. 


13,150 


50 


etc. 


etc. 


etc. 


etc. 


etc. 


etc. 


etc. 


etc. 



Notes 
Operating the above projector lamps at over-voltage will result in the following: 



Per Cent 
Volts 

100 
105 

110 
115 
120 



Per Cent 
Lumens 

100 
120 
140 
160 

200 



Per Cent 
Life 


Per Cent Increase 
in Rate of Lumen 
Depreciation 


Probable 
Destruction of 
Bulb Lamps 


100 







50 


5 




25 


8 


A, E, etc. 


12 


11 


D, G, etc. 


10 


15 


F, etc. 



Projector Classification 

When used with the standard lamps commercially recommended for the particular 
projector and tested at the rated lumen output of the lamps, projectors fall into 
the following classification: 



Class A 
Class B 
Class C 



800-1000 lumens 
600- 800 lumens 
400- 800 lumens 
etc. 



204 F. E. CARLSON [J. S. M. P. E. 

Foot-Candle Intensity on Screens of Various Sizes 

Lumens Av. Foot-Candles 

4 Ft. 6 Ft. 8 Ft. 10 Ft. 12 Ft. 14 Ft. 16 Ft. 
80 7 5 

100 9 8 

200 15 5 1.5 

400 8 5 3.7 

600 5.5 4 3.1 

1000 7 5.2 

It seems to me that if the Committee on Non-Theatrical Equipment should 
develop something like these tables it would be a contribution of tremendous 
value. 

MR. MITCHELL: Any attempt to measure illumination at the projection aper- 
ture is objectionable for three principal reasons: first, it does not take into ac- 
count the efficiency of the projection lens; second, due to the small area of the 
aperture, the chance of error is too great; third, and perhaps most important, 
it does not enable the distribution of light across the field to be checked. 

It is therefore recommended that the illumination be measured on a screen. 
Inasmuch as the only thing that really counts is the efficiency of the projector 
while showing pictures, all measurements should be taken with the projector 
running, preferably at not less than the normal speed of sixteen pictures a second. 
A standard screen size should be determined a screen 3 feet wide would be large 
enough to obtain accurate measurements from the corners, edges, and center to 
check the illumination intensity distribution. The 3-ft. screen is a very popular 
size, and is a convenient standard. 

A standard meter such as the new Weston photronic meter might be preferred 
on account of the better reproducibility of measurements. Nine readings might 
be made on such a screen: namely, at the four corners; at the top, bottom, and 
both sides; and at the center. The average of the nine readings could be taken 
and from that figuring the size of screen used the effective lumen output could 
be determined. In making such a test, it would be essential to use lamps of a 
uniform color temperature rating for lamps nominally the same. These would 
then be burned in the projector under exact voltage conditions so that fair com- 
parative readings might be obtained. 

Reverting to the matter of illumination distribution over the screen, it is recom- 
mended that variation in illumination from the corners or edges to the center 
should not exceed 15 per cent. The reason for this is that unless the projection 
optics including the centering of the lamp, etc. are correctly set, there might be 
a hot spot at the center of the screen that would upset any attempt to obtain a 
reasonable comparative reading. The suggestion of averaging nine readings 
has in mind the elimination as far as possible of any variation of this sort. 

MR. CARLSON: The question was asked about the effect of operating in- 
candescent lamps at other than their normal voltages or currents. Life changes 
according to about the 13th power of the voltage, light output to about the 3.5 
power, and watts to about the 1.5 power. Since source size remains constant, 
wattage per unit area and source brightness change directly in proportion to 



Mar., 1935] LlGHT SOURCES FOR PROJECTION 205 

changes in wattage and light output, respectively. Operating a lamp above its 
design voltage results in slightly less total blackening but, since lamp life is short- 
ened by over-voltage, the rate of blackening is correspondingly more rapid. 
The reverse is true when the lamp is operated at less than its design voltage. 

It has been suggested that further advances in screen illumination might be 
made with professional equipment if sufficient forced ventilation were provided to 
permit high wattages in relatively small bulb sizes. It is a fact that manu- 
facturers of 16-mm. projectors have gone further in this direction than have the 
manufacturers of any of the various types of 35-mm. projectors. Probably much 
can still be done along these lines. 

The subject of projector ratings or classifications is so far-reaching and presents 
so many complications that President Goldsmith has done well to refer the 
matter to the Non-Theatrical Equipment Committee in lieu of further discussion. 
Mr. Farnham has already pointed out that the developments of the past few years 
were particularly favorable to creating confusion hi the minds of consumers and 
that we are now arriving at a point where this condition will doubtless be less 
marked. Mr. Porter mentioned the procedure employed with health lamps 
using ultra-violet sources. It may be well to point out that in that case there are 
three governing factors that have no parallel in the projector field: namely, 
there is a potential hazard to the user which makes specific instructions impera- 
tive; second, one is dealing with radiation that is not visible nor otherwise makes 
itself evident to the senses until after use; third, the technic of measurement is 
such that very, very few laboratories are equipped to function in this field. The 
reasons for a central rating organization are considerably less compelling in the 
case of projectors and, on the other hand, the many variables involved present so 
many more complications that any system adopted would have to be rather ar- 
bitrary and thus be more difficult of satisfactory administration. This further 
point must be kept in mind: despite any rating scheme, visual demonstration 
will doubtless continue to be the chief selling method. Abuses are not un- 
known, but the trend in the industry is definitely toward fair practice in that 
regard. 



A MECHANICAL DEMONSTRATION OF THE PROPERTIES 
OF WAVE FILTERS* 

C. E. LANE** 



Summary. The nature of wave motion and amplitude and phase distortion of 
waves are discussed. Wave filters of different types and important uses of wave 
filters are defined and discussed, and finally, a mechanical model of a wave filter 
is described to demonstrate visibly their essential characteristics. 

Much of our experience is concerned with wave motion, i. e., vibra- 
tory motion which originates at some source and which travels away 
from that source. These waves may occur in solid objects, in liquids, 
or in gases. An example of the last is sound waves traveling in air. 
Wave motion may occur also in wires carrying electric currents, the 
motion being the displacement of the electrons in the conducting 
wires. If wave motions are slow, corresponding to only a few oscilla- 
tions per second, they are said to be of low frequency. If the waves 
are rapid they are said to be of high frequency. 

The simplest kind of wave motion is simple harmonic motion, like 
that observed when a pendulum swings freely back and forth. The 
back-and-forth motion of the air particles when a pure tone is trans- 
mitted through the air is like the motion of the swinging pendulum. 
Such a simple wave motion is said to be of a single frequency, and 
when the displacement or the velocity of the motion is plotted as a 
function of time, the resultant curve is the familiar sine or cosine 
curve. Most wave motions with which we are concerned are complex, 
being made up of a large number of superimposed simple-harmonic 
components, or single-frequency waves which differ in frequency and 
magnitude. As a rule, the components of complex waves bear a 
simple harmonic relation one to the other, there being a fundamental 
frequency of vibration, and the frequencies of all of the other compo- 
nents are multiples of the frequency of the fundamental. Speech and 
music are complex waves made up of simple wave components, and 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 
** Bell Telephone Laboratories, Inc., New York, N. Y. 
206 



DEMONSTRATION OF WAVE FILTERS 



207 



the important components are of frequencies ranging from about 60 
or more vibrations per second upward as high in some cases as 8000 
or 10,000 per second. 

Curves A, A', B, and C of Fig. 1 are simple waves. The ordinates 





SIMPLE 

COMPONENT 

WAVES 



COMPLEX 
WAVE 



ATTENUATION 
"DISTORTION 



TIME-* 

FIG. 1. Simple waves and attenuation distortion of complex waves. 



A SHIFTED 30*\ 



v / 




ORIGINAL 

-COMPLEX 

WAVE 



DISTORTION 



TIME- 
FIG. 2. Delay and phase-distortion of complex waves. 



208 C. E. LANE [j. s. M. P. E. 

represent amplitude, velocity, or current, as the case may be, and the 
abscissas represent time. The two lower curves are the complex 
waves resulting from the simultaneous occurrence of the three simple 
components. Except for modulation, which is the introduction of 
new frequency components not originally present, there are but two 
things that can happen to change the shape of a complex wave as it 
appears at different places. The first of these is a change in the rela- 
tive magnitude of the components, generally called attenuation distor- 
tion. The difference between the two lower curves of Fig. 1 illus- 
trates attenuation distortion. The other change which may take 
place arises from phase-shift, or the difference in time at which the 
component waves pass through the maxima or minima values. When 
the phase-shift of all the component waves, measured in degrees or 
radians (one complete period is 360 or 2ir radians), is proportional to 
the frequency, there is no phase distortion but a delay given in sec- 
onds by the slope of the phase curve d-B/dco, where B is in radians and 
w is 2r times the frequency in vibrations per second. Delay without 
distortion is illustrated by the middle curve of Fig. 2. When the 
phase-shift is not proportional to the frequency, the shape of the com- 
plex wave is changed. This is illustrated by the lowest curve shown 
in Fig. 2. 

Wave filters are devices which may be put in the paths of these 
wave motions and which will virtually stop waves of some frequencies 
and allow waves of other frequencies to pass through them practically 
unchanged in magnitude. Wave filters may be made of solid parts 
and put in the paths of waves traveling through solid objects, or they 
may be made in the paths of sound waves in the air; or they may be 
made of electrical parts and put in the paths of electrical waves travel- 
ing through wires. 

Filters which will permit waves of all frequencies up to some definite 
desired frequency to pass through them unaffected, but greatly at- 
tenuate waves of higher frequencies, are called low-pass filters. Filters 
which will stop the lower frequencies and permit the higher frequen- 
cies to go through them are called high-pass filters ; and filters which 
will pass a band of frequencies over a certain definite frequency range 
and attenuate frequencies above and below this range are band-pass 
filters. 

Electrical wave filters are used extensively in telephone systems to 
eliminate disturbance waves which otherwise would appear along 
with the transmitted message waves and interfere with the message 



Mar., 1935] 



DEMONSTRATION OF WAVE FILTERS 



209 



reception. They are used to some extent for this same purpose in 
recording and reproducing circuits of sound-picture systems. In 
carrier-current telephony and telegraphy, for the purpose of economy, 
a number of messages are often transmitted simultaneously in differ- 
ent frequency ranges over the same wires. Here filters are used to 
separate the different messages, on the basis of frequency content, and 
confine them to their assigned frequency channels. 

The illustrations that follow are photographs of a large mechanical 
filter* which can be used to demonstrate the action of wave filters. 
Although the filter is a particular type of band-filter, it possesses many 
features that are common to all types of filters. A mechanical filter 
is used for this purpose in preference to an electrical filter so that one 
can see the effect of the filter upon the transmitted wave. The filter 
is made large and its band of transmission is at low frequencies so that 
the motion can be followed both for frequencies in the transmission 
band and for frequencies adjacent to the transmitted band which are 




FIG. 3. Electrical analogue of mechanical filter shown in Fig. 4. 

stopped by the filter. The filter consists of a number of pendulums 
connected together by flexible springs, and is so designed that it will 
permit wave motions of frequencies falling between 45** vibrations 
per minute and 59 vibrations per minute to pass through it but will 
greatly attentuate frequencies below or above this range. The filters 
used in voice- and carrier-frequency telephone circuits generally oper- 
ate at frequencies which are from 1000 to 20,000 times as high as this. 
If this filter were made of electrical parts instead of mechanical 
parts, its schematic diagram would be as shown in Fig. 3, which is its 
"electrical analogue." The mass of the pendulum bobs corresponds 
to the series inductances, the attraction of gravity upon the bobs corre- 

* Except for small modifications this filter is the same as that described in the 
J.A.I.E.E., Dec., 1933. 

'* Because of the very low frequency it is described in vibrations per minute 
instead of per second as is usual. 



210 



C. E. LANE 



[J. S. M. P. E. 




FIG. 4. Mechanical band-pass filter, with transmission band between 45 
and 59 vibrations per minute : at rest. 




FIG. 5. 



Operation of mechanical filter below lower cut-off: 44 vibrations 
per minute. 




FIG. 6. 



Filter operating in transmission band just above lower cut-off: 
46 vibrations per minute. 



Mar., 1935] DEMONSTRATION OF WAVE FILTERS 211 




FIG. 7. Middle of transmission band of filter: 52 l /z vibrations per minute. 




FIG. 8. Just below upper cut-off of filter: 58 vibrations per minute. 




FIG. 9. Just above the upper cut-off of filter: 60 vibrations per minute. 



212 C. E. LANE [J. S. M. P. E. 

spends to the series capacitances, and the arched springs connecting 
the adjacent pendulums provide the shunt capacitances. In this 
comparison the velocity of motion in the mechanical filter corre- 
sponds to the current in the electrical filter. The first pendulum of the 
filter on the left may be driven sinusoidally by the flywheel of a small 
motor through the driving shaft and connecting spring shown in 
Fig. 4. A resistance termination is provided for each end of the filter 
by allowing the two bobs at the ends of the filter to swing through vis- 
cous oil. 

The principal property of wave filters is that of permitting to pass 
through them wave motions of some frequencies and greatly attenuat- 
ing waves of other frequencies.* The difference in frequency be- 
tween waves freely transmitted and those not transmitted may some- 
times be very small. In order to show this property of wave filters a 
number of photographs have been taken of the filter in operation. 
These photographs for different frequencies are shown in Figs. 5 to 9, 
inclusive.** In the case of Fig. 5 the frequency of the applied wave 
is 44 vibrations a minute. For this frequency the amplitude or veloc- 
ity of motion of each successive pendulum bob is about half that of 
the previous bob. The motion transmitted through the filter to the 
far end is less than one hundredth of the motion at the end of the filter 
at which the force is applied. (The attenuation at this frequency is 
over 40 db., as expressed in terms of the standard logarithmic unit for 
measuring transmission.) Fig. 6 is for a frequency of 46 vibrations a 
minute, only a little over 4 per cent faster than shown in Fig. 5. How- 
ever, there is now a decided change in the transmission. At this fre- 
quency there is very little change in the amount of motion from one 
successive pendulum bob to the next, and for the entire filter the 
amplitude of motion of the last pendulum is about three-fifths the 
amplitude at the input end where the motion starts (about 4 db. 
attentuation). Fig. 7 is for a frequency of 52 x /2 vibrations a minute, 

"The primary object of wave filters is effectively to eliminate certain frequency 
components. A device for changing the relative magnitude of desirable wave 
components to compensate for attentuation distortion is called an attentuation 
equalizer. 

** All the photographs were taken after sufficient time had elapsed to allow the 
filter to reach a steady-state condition. For this filter, since it is working at such 
a low frequency, the time required may be a minute or more. The time required 
for the same type of filter for a mid-band at 10,000 cycles per second would be 
about 0.005 second. 



r., 1935] 



DEMONSTRATION OF WAVE FILTERS 



213 



rhich is just about the middle of the transmitting band of the filter. 

it this frequency there is no diminution whatever in the amplitude of 
)tion due to its transmission through the filter. Fig. 8 is for a fre- 
quency of 58 vibrations a minute, which is just about the highest fre- 
quency that the filter will transmit. Fig. 9 is for a frequency of 60 
vibrations per minute. At this frequency the amplitude is reduced 
by a factor of about two for each filter section, or from one pendulum 




50 55 

FREQUENCY IN R.P.M. 



FIG. 10. Attenuation of filter in decibels, as a function of frequency. 

bob to the next, again giving a total attenuation of over 40 db. At 
any frequency lower than 45 vibrations per minute or higher than 60 
vibrations per minute the effectiveness of the filter in stopping the 
wave is greater than at frequencies between these values. Fig. 10 is 
a plot in decibels of the attenuation of the filter as a function of the 
frequency. 

The photographs shown in Figs. 5 to 9 have been taken in such a 
way as to show more than the attentuation of the filter. The photo- 



214 



C. E. LANE 



[J. S. M. P. E. 



graphic exposure was maintained for several seconds so as to bring out 
the total arc of the swing of each pendulum. However, at some in- 
stant during the exposure an instantaneous flash was made, so as to 
record the positions of all the pendulums at that instant and thus to 
show the phase-shift. Phase-shift refers to the difference in the time 
of occurrence of maximum displacement (or any other point of refer- 
ence) of the wave in different parts of the filter. In Fig. 5 all the 
pendulum bobs are moving together; that is, they all pass through 



HI 140 



g. 00 

S 

t 8 
1 
81 eo 



20 



PHASE SHIFT = X 360 



50 55 

FREQUENCY IN R.P.M. 



FIG. 11. Phase-shift per filter section, in degrees, as a function of 
frequency. 

zero at the same time, and all have maximum displacements at the 
same time. Hence all the pendulum bobs are in phase, and there is no 
phase-shift. In Fig. 6 all the pendulum bobs are moving almost to- 
gether, but not quite. The bob at the extreme right is about one- 
quarter of a complete period behind in its motion as compared with 
the one at the extreme left. The phase-shift from one successive 
pendulum bob to the next, or the phase shift per section of the filter, 
is about 12 degrees. In Fig. 7, every other bob is exactly opposite in 



Mar., 1935] DEMONSTRATION OF WAVE FILTERS 215 

phase, or 180 degrees apart. Hence there is at this frequency a 90- 
degree phase-shift per filter section. In Fig. 8, the phase-shift per 
section is about 150 degrees; that is, the motion of adjacent pendu- 
lum bobs is almost opposite in phase, but not quite. In Fig. 9 
there is a ISO-degree phase-shift in passing from one pendulum to 
the next. 

We have noticed, then, that for the frequency of 44 vibrations per 
minute there is no phase-shift in the filter. This condition holds for 




FIG. 12. The delay in the passage of transient impulses through the filter. 
( Upper) transient just starting; (middle) after 4 sec., transient approaching 
midpoint; (lower) after 10 sec.; transient reaching end of filter. 

all frequencies below the transmitting range of the filter. We have 
also noticed that for 60 vibrations per minute there is a 180-degree 
phase-shift per section of the filter. This condition holds for all fre- 
quencies above the transmitting range of the filter. Within the trans- 
mitting band of the filter the phase-shift changes gradually from to 
180 degrees per section, passing through 90 degrees at about the mid- 
dle of the band of the filter. A plot of this phase-shift per filter sec- 
tion as a function of frequency is shown in Fig. 11. The phase-shift 
characteristic shown is for this particular type of band filter. How- 



216 C. E. LANE [j. s. M. p. E. 

ever, it may be stated as a general proposition that for any filter what- 
ever the phase-shift in the non-transmitting ranges remains constant 
with frequency change either at degrees or 180 degrees, whereas in 
the transmitting range there is a continuous increase in the phase- 
shift with increasing frequency. 

We see, then, since the phase-shift of filters in general is not propor- 
tional to the frequency, that they introduce phase distortion. Fortu- 
nately, however, this distortion does not generally affect the quality 
of speech or music waves transmitted through them. The ear can not 
detect ordinary changes in the relative phases of the components of 
such complex waves. It would take fifty or more filters of the usual 
type in tandem to produce a detectable phase distortion of speech or 
music waves transmitted through them. 

There is an initial delay in the attainment of the steady-state 
condition required for the steady transmission of disturbances through 
a filter, which is practically independent of the nature of the wave 
being set up. This initial delay can be related to the phase charac- 
teristic of the filter. It is determined by computing the minimum 
slope of the phase characteristic in its frequency range of transmis- 
sion. For this filter the delay is about 9 or 10 seconds. What hap- 
pens after this initial delay depends upon the nature of the wave 
being started. 

The photograph shown in Fig. 12 was taken to illustrate this delay. 
The upper photograph shows the motion taking place in the filter 
during the first half second after starting a disturbance at the input 
of the filter by displacing the first pendulum bob and suddenly releas- 
ing it. During this interval the motion is confined solely to the first 
section of the filter. The middle photograph was taken between the 
fourth and fifth second after starting the disturbance. By this time 
this disturbance has reached the middle of the filter. The lower 
photograph shows the interval between the ninth and tenth second 
after starting the disturbance, by which time the disturbance has been 
propagated entirely through the filter. In other words, the delay of 
the filter due to the transient disturbance was between 9 and 10 sec- 
onds. Had the disturbance been caused by suddenly starting the 
driving motor at any frequency, the initial delay would have been the 
same. 

In order to illustrate the effect of reflection which occurs when fil- 
ters are improperly terminated, the photograph shown in Fig. 13 was 
taken. This is for a frequency of 52 x /2 cycles per minute, for which 



Mar., 1935] 



DEMONSTRATION OF WAVE FILTERS 



217 



the phase shift is 90 degrees per section. Before taking this photo- 
graph, the terminating impedance of the filter was removed; that is, 
in language applicable to the electrical analogue of the filter, the filter 
was short-circuited at the output end. This caused complete reflec- 
tion of the wave at the end of the filter. It will be noticed that the 
amplitude of every evenly numbered pendulum bob is doubled due to 
this reflection, the reflected wave being exactly in phase with the di- 
rect wave at these positions. On the other hand the reflected wave is 
approximately 180 degrees out of phase at the oddly numbered pendu- 
lum bobs, and hence the direct and reflected waves nearly cancel each 
other at these points and the motion is very small. If the frequency 




FIG. 13. Illustrating reflection occurring when filter is improperly 
terminated; in this case a short circuit: frequency, 52 x /2 vibrations per 
minute. 



were to remain the same and the output of the filter open-circuited by 
holding the last bob still instead of short-circuited, as in Fig. 13, 
there would still be complete reflection at the end of the filter. In this 
case, however, the nodes would appear at the evenly numbered pendu- 
lum bobs instead of at the oddly numbered ones, as shown in the 
photograph. Photographs of this kind might be taken at other fre- 
quencies, and the resultant motion at any point along the filter would 
depend upon the relative phases of the direct and reflected waves. 

The preceding discussion has been restricted largely to a certain 
type of band-pass filter. There are many other types of band-pass 
filters, the element configurations of which differ considerably. Also 
high-pass and low-pass filters of various element arrangements are 



218 C. E. LANE [J. s. M. P. E. 

used. These filters are described in many different publications.* 
The most complete description is perhaps by T. E. Shea, with tables 
showing the configurations for all the most commonly used filter sec- 
tions and their attentuation and phase characteristics. Formulas are 
also given for the values of the elements of the various filter sections. 
Complete filters are formed by joining together a number of filter sec- 
tions which may be alike, as in the case of the filter demonstrated, or 
which may differ from each other. In the case where filter sections of 
different types are connected together, it is necessary that the impe- 
dances of the sections at their junctions be alike. The total attenua- 
tion and phase characteristics of a composite filter is the sum of the 
characteristics of the individual sections which go to make up the 
filter. 

BIBLIOGRAPHY 

SHEA, T. E.: "Transmission Networks and Wave Filters," D. Van Nostrand 
Co., New York, N. Y., 1929. 

JOHNSON, K. S. : "Transmission Circuits for Telephonic Communication," 
D. Van Nostrand Co., New York, N. Y., 1927. 

COLPITTS, E. H., and BLACKWELL, O. B.: "Carrier Current Telephony and 
Telegraphy," Trans. A.I.E.E., 40 (Feb., 1921), No. 2, p. 205. 

HAMILTON, B. P., NYQUIST, H., and LONG, M. B., and PHELPS, W. A.: "Voice 
Frequency Carrier Telegraph System for Cables," /. A.I.E.E., 44 (March, 1925), 
No. 3, p. 213. 

HARTLEY, R. V. L.: "The Transmission Unit," Elec. Comnt., 3 (July, 1924), 
No. 1, p. 34. 

JOHNSON, K. S., AND SHEA, T. E.: "Mutual Inductance in Wave Filters, with 
an Introduction on Filter Design," Bell Syst. Tech. J., 4 (Jan., 1925), No. 1, p. 52. 

STEWART, G. W.: "Acoustic Wave Filters," Phys. Rev., 20 (Dec., 1922), No. 
12, p. 528. 

ZOBEL, O. J. : "Distortion Correction in Electrical Circuits with Constant Re- 
sistance Recurrent Networks," Bell Syst. Tech. J., 7 (July, 1928), No. 3, p. 438. 

ZOBEL, O. J.: "Theory and Design of Uniform and Composite Electric Wave 
Filters," Bell Syst. Tech J., 2 (Jan., 1923), No. 1, p. 1. 

ZOBEL, O. J.: "Transmission Characteristics of Electric Wave Filters," Bell 
Syst. Tech. J., 3 (Oct., 1924), No. 4, p. 567. 

LANE, C. E.: "Phase Distortion in Telephone Apparatus," Bell Syst. Tech. J., 
9 (July, 1930), No. 3, p. 493. 

CLEMENT, A. W.: "Line Filter for Program System," Elect. Eng., 53 (Apr., 
1934), No. 4, p. 562. 

PAYNE, E. B. : "Impedance Correction of Wave Filters," Bell Syst. Tech. J. t 9 
(Oct., 1930), No. 4, p. 770. 

BODE, H. W.: "A Method of Impedance Correction," Bell Syst. Tech. J.. 9 
(Oct., 1930), No. 4, p. 794. 

* See bibliography. 



Mar., 1935] DEMONSTRATION OF WAVE FILTERS 219 

DISCUSSION 

MEMBER: How serious is the phase distortion in electrical filters, in general? 
How are some of the other filters designed? 

MR. LANE: A band of frequencies, roughly 3000 cycles wide, is required for 
reasonably good transmission of speech or music. This band may be located in 
the voice-frequency range or, by modulation, may be located at any position 
desired above this range. Filters are used in transmitting speech or music over 
telephone circuits which will pass the frequencies desired and provide 40 or 50 db. 
attenuation to other frequencies. The speech or music may pass through a 
number of filters in tandem and the effect of phase distortion is not detectable. 
Roughly speaking, it would take 40 or 50 ordinary band filters, providing this 
amount of discrimination, connected in tandem, to produce a phase distortion that 
would be at all noticeable. Only in very special cases is it necessary to use so 
many filters together, and when it is necessary, special filter designs are available 
that will produce less phase distortion than the ordinary and less expensive types. 

As you observe, this filter may be regarded as built up of seven identical sec- 
tions in tandem, each section having an inductance and condenser in the series 
arm and a single condenser in the shunt arm. This is one type of band-pass filter. 
The simplest filters are low-pass and high-pass filters. A low-pass filter has an 
inductance in the series arm and a condenser in the shunt arm of the filter 
section; and a high-pass filter, a condenser in the series arm and an inductance 
in the shunt arm. A low-pass filter such as this one will pass frequencies with 
little or no loss up to some certain frequency and from there on, the loss at higher 
frequencies will increase rather gradually, becoming infinite at infinite frequency. 
The simple high-pass filter has infinite loss at zero frequency, and the loss de- 
creases gradually as the transmitting range of the filter is approached. 

Both the low-pass filter and the high-pass filter sections may be modified so as 
to provide more abrupt increase in the attenuation in passing from the transmit- 
ting range to the attenuating range of the filter. This is accomplished in the low- 
pass filter by the use of a condenser in parallel with the inductance in the series 
arm or by use of an inductance in series with the condenser in the shunt arm ; and 
in the high-pass filter by the use either of an inductance in parallel with the con- 
denser of the series arm or a condenser in series with the inductance in the shunt 
arm. High-pass and low-pass filters having these additional elements have peaks 
of infinite attenuation which may be located as near or as far away from the cut- 
offs of the filter as desired. The closer these attenuation peaks are located to the 
edges of the transmitting ranges of the filters, the more abrupt will be the dis- 
crimination at the cut-offs of the filter. However, the movement of the attenua- 
tion peaks toward the cut-offs of the filter lowers the loss per filter section beyond 
the attenuation peaks. 

The band-pass filter of the type of this model may be given an attenuation 
peak on the upper side of the band by introducing an inductance in series with 
the condenser in the shunt arm and modifying the value of the condenser; or by 
a condenser in parallel with the inductance and condenser of the series arm. 
The type of band filter used in this demonstration produces more loss above the 
transmitting band than below. There is a type of band filter available for use 
which is symmetrical in loss about the transmitting band. This filter has an 
inductance and condenser in series for the series arm, and an inductance and 
condenser in parallel for the shunt arm. 



220 C. E. LANE 

MEMBER: What would happen if there were no oil at the end of of the filter? 

If there were no oil, it would be equivalent to terminating the filter in a 
short-circuit; as can be seen from the fact that if I should remove the oil, a little 
at a time, so that the pendulum should make less and less contact with it, the 
value of the terminating resistance would become smaller and smaller. The filter 
is open-circuited when the terminating impedance is very high ; that is, it is open 
circuited when I held the last pendulum still with my hand. If the filter were 
short-circuited, or the oil removed, maximum motion would occur at the last 
pendulum bob and a node would occur at the one adjacent to it; the effect would 
be just the opposite of open circuiting the filter. 

Mr. SHEA : Notice that it is the velocity of the balls that corresponds to current 
through an inductance and not the displacement of the balls; so when thinking 
of what an open-circuit and a short-circuit is like in a mechanical system, you 
must consider the velocity and not the displacement. 

Mr. McMANN: The behavior of an electrical filter has been shown here by 
means of a mechanical model. Isn't it so that complex mechanical problems are 
sometimes transformed into electrical analogues for solution, as in the case of loud 
speakers, microphones, etc., with their masses and compliances? 

MR. LANE: That is quite true. Electrical analogues are very frequently 
used which correspond to some mechanical system, like the loud speaker or the 
ear. To aid in understanding the performance, such electrical analogues are 
quite useful, though generally only approximate. They can be readily solved 
and understood, and afford good approximate solutions for the performance of 
mechanical systems. 

MR. SHEA: Basically this is a demonstration of energy transformations. The 
relation between electrical and mechanical movement and the principles that 
underlie the mechanical wave filter are very much the same as underlie the 
loud speaker, where the air is pushed around instead of oil. This applies also 
to microphones, phonograph pick-ups, phonograph recorders, light valves, oscil- 
lographs, etc. 

Some one several years ago pointed out that between the time the sound reached 
the microphone in the studio and the time it issued from a loud speaker in a 
theater, there might be as many as forty transformations from one kind of energy 
to another. If you were to pick out one course of education above all others that 
a motion picture engineer ought to go through in school, it would be on the rela- 
tionships between wave motions in different forms electrical, mechanical, 
magnetic, optical, and so forth because in motion pictures you do go from one 
condition to another many times; even in a single projector installation. And 
that is the reason why this demonstration is so significant here, not as illustrating 
wave niters so much, but as illustrating the very things with which we deal every 
day. 



ELECTRONIC TUBE CONTROL FOR THEATER LIGHTING* 
J. R. MANHEIMER AND T. H. JOSEPH** 

Summary Modern theatrical practice requires a switchboard capable of con- 
trolling elaborate electrical effects in connection with stage spectacles. Three-color 
house lighting, four- and five-color stage effects, frequently involving as much power as 
1500 kw., require a flexible, compact, easily controlled system for accurately and 
rapidly effecting the various combinations and changes of lighting. 

The reactance dimming electronic tube controlled switchboard here described is 
capable of presetting the intensities and combinations of lights, and easily controlling 
the effects required for several scenes in advance without interfering with the combina- 
tions for the scene in use. 

Rectifier-tube control for stage-lighting systems may be classed as a 
rather recent development. Naturally, the question has arisen as to 
why such a system is used. The answer is simple. 

First, due to the increased size of the newer theaters, the number of 
circuits to be controlled and their wattages have increased to such an 
extent that the resistance type of dimmer has become impracticable 
in many cases. The resistance dimmer needs considerable contact 
pressure to carry its load and, as a result, the muscular effort required 
to operate a large bank of dimmers makes it impracticable. Second, 
stage space is always at a premium, making it desirable as well as 
economical to locate the dimmers at a remote point and to control 
them from a pilot-board at stage level. This, of course, might be 
done with motor-operated dimmers, but with such a system the 
flexibility afforded by a rectifier tube controlled system can not be 
achieved. Furthermore, the maintenance of such a system is an 
endless and expensive task. Third, for a succession of rapid light 
changes, presetting of resistance type dimmers for each change after 
the first is impracticable, but is conveniently accomplished with a 
tube-controlled board. Toward the end of attaining a stage-lighting 
system that is compact, easy to operate, and economical to maintain, 
the development of the tube type of stage switchboard has been 
aimed. 






* Presented at the Fall, 1934, Meeting at New York, N. Y. 
** E-J Electric Installation Co., New York, N. Y. 

221 



222 



J. R. MANHEIMER AND T. H. JOSEPH [j. s. M. p. 



The so-called preset switchboards providing only for the presetting 
of circuits have long been familiar. The real value of presetting lies in 
being also able to preset dimming. In order to do so, it is necessary 
to have a control unit so small that five or more can be mounted in a 
space much smaller than that originally occupied by a single control 
on the old type of preset board. Only tube control has made this 
possible; hence the development in this line. As the development 
advanced, it was found that other desirable features such as propor- 




FIG. 1. 



An elementary reactance dimmer- 
control system. 



tional fading, compact master control, and efficient remote control 
could easily be realized. 

Although several types of tube controlled theater boards are in use 
at the present time, this paper will restrict itself to a description of 
one of the latest of these types which has just been put into operation 
in the Center Theater at Radio City, New York, N. Y. 

In order to describe the system and explain wherein it differs from 
other tube control systems, let us first consider an elementary reac- 
tance dimmer-control system such as shown in Fig. 1. This consists 
of a reactance dimmer in series with the lamps which are connected 
across the a-c. power supply. The dimmer reactor consists of a 



Mar., 1935] TUBE CONTROL FOR THEATER LIGHTING 223 

three-legged saturable core reactor having an a-c. coil wound upon 
each outside leg and a d-c. coil upon the center leg. The two a-c. 
coils are connected in series with each other and with the lamps. The 
d-c. coil is connected to a variable source of direct current. 

With no excitation in the d-c. coil, the impedance of the two a-c. 
coils is such as to cause a voltage drop in their windings sufficient to 
limit the lamp voltage to the desired minimum. The connections of 
the coils are such that in the center leg of the reactor the a-c. flux of each 
coil at any instant is equal and opposite to the flux of the other coil. 
Therefore, no alternating voltage is induced in the d-c. coil. When 
the latter is energized to the point of saturation of the iron in the 
reactor, the a-c. coils are unable to induce a voltage in their own 
windings, and the lamp voltage is equal to the line voltage less the 
drop due to the resistance in the a-c. coils. Intermediate values of 
lamp voltage are attained by intermediate values of direct current in 
the d-c. coil. 

The intensity of light is varied by changing the small amount of 
direct current passing through the d-c. coil, "full bright" being at- 
tained when the direct current is at its maximum and "black out" 
when at its minimum. In order to burn the lamps at full brilliancy 
when the reactor is saturated, it is necessary to provide means of 
compensating for the voltage drop due to the resistance of the reactor 
windings. This is accomplished by a small booster transformer hav- 
ing a secondary voltage equal to the drop in the reactor and connected 
in series with the lamp circuit. The primary of this transformer may 
be connected to any source of alternating current of the proper volt- 
age. 

Reference to the diagram of the tube-controlled circuit in Fig. 2 
will immediately suggest a great similarity to the circuit previously 
described. In the first place, the reactance dimmer is the same. 
The source of direct current consists of a full-wave rectifier tube ; but 
the variable resistance for controlling this direct current has been 
replaced by the "hysterset," which is a device for enabling small 
amounts of power to control large amounts of power. The "hyster- 
set" is controlled by a variable resistor connected across its low- volt- 
age control circuit. Referring to Fig. 2, its mode of application is as 
follows : 

Starting at the pilot switchboard, two transformers, of 120/12 volts, 
connected across two phases of a three-phase supply, provide the 
proper supply voltages. Across the outer ends of this supply is con- 



224 



J. R. MANHEIMER AND T. H. JOSEPH [J. S. M. P. E. 



nected the control resistor, which has an adjustable slider. A small 
copper-oxide rectifier is used, having one anode connected to the ad- 
justable slider and the other 4o one side of the transformer secondary. 
The cathode circuit of the rectifier is connected through the control 
coil of the hysterset to the middle point of the secondary. The pur- 



120/I2V CONTROL TRANSFOCMEJJ. 



5 VARIABLE RESISTOR CONTBOU 




-o- 
-o 
o- 

FIG. 2. The tube -controlled circuit employing the hysterest. 

pose of this is to supply to the hysterset direct current which can be 
controlled as follows : 

When the slider is in the lowest position, both anodes of the rectifier 
are connected to the same source, so that there is no difference of 
potential or phase angle between them and the rectifier acts as a half- 
wave rectifier with the anodes in parallel. The reactance of the con- 
trol coil then permits only a minimum amount of current to flow, 
corresponding to the "black out' ' position of the dimmer. If the slider 
is moved to the other end of the resistor, the rectifier acts as a full- 



Mar., 1935 J TUBE CONTROL FOR THEATER LIGHTING 



225 



wave rectifier with the anodes 120 degrees out of phase. This allows 
maximum current to flow in the control coil, corresponding to the 
"full bright" position of the dimmer. Intermediate light intensities 
are attained at corresponding intermediate settings of the resistor. 
The control coil operates through the hysterset so that during each 




FIG. 3. 



Front view of the tube-controlled pilot board in the Center 
Theater, New York, N. Y. 



half cycle the iron of the anode reactor is conditioned so as to prede- 
termine the amount of current to flow during the next half -cycle of 
operation. The output of the hysterset is fed into a full- wave recti- 
fier tube, the cathode of which is connected through the d-c. coil of 
the dimmer reactor back to the mid-point of the anode transformer. 
Thus by varying the pilot circuit, as previously described, the direct 



226 



J. R. MANHEIMER AND T. H. JOSEPH [J. S. M. P. E. 



current in the d-c. coil is varied and the voltage at the lamps changed 
accordingly, as illustrated in Fig. 1. 

Fig. 3 is a front view of the tube-controlled pilot board in the 
Center Theater. The individual sections at the left consist of one 
hundred and thirty stage controls, each with five presets and a re- 
hearsal section, arranged according to color: top row, amber; next, 
red, green, and blue. On the right are fifty-one similar control units 
for the house lights. 

The top row of thirty-two quadrants with handles control Selsyn 
generators operating four color frames in every incandescent spot and 
flood on the stage and in the house. The row below this, in the center 




FIG. 4. Part of one of the attic reactor racks, with the 
reactors in place. 

section, contains the Selsyn color masters and their grand master. 
Below these are smaller quadrants used as scene masters, five for 
stage and five for house. In the lowest row are the supplementary 
scene masters, five for stage and five for house lights. Between the 
row of scene master controls and supplementary controls are two 
buttons. The button at the left is used for "black out stage" and the 
button at the right for "black out house." 

At each side of the center section are the color masters and stage 
and house grand masters. The two large handles at each side of the 
center panel are the stage and house faders. Immediately above each 
is a bank of five pairs of interlocking push-buttons for selective fading 



Mar., 1935] 



TUBE CONTROL FOR THEATER LIGHTING 



227 



from one scene to any other preset scene. Scenes may be faded one 
into the other in any sequence or combination. Immediately at the 
center is a lock that shuts off the entire system except the work-light 
switches at the lower right. At the left and across the bottom left 
are individual controls and group masters to control pockets in the 
stage floor and elsewhere. At the upper right is a guarded "panic 
light" switch, which in conjunction with two others in the house, 
throws on the "full bright" amber house lights regardless of the posi- 
tion of the dimming controls. 

Fig. 4 shows part of one of the attic reactor racks with the reactors 
in place. There are two reactor rooms, one in the basement and one 
on the gridiron level, so as to 
shorten the lengths of the circuits 
and so reduce the voltage drop. 
Circuits supplying the footlights, 
pockets, proscenium and portal 
floods, tower spots, etc., are con- 
nected to the basement reactor 
group. Circuits running to bor- 
ders, all top lights, and auditor- 
ium ceiling are connected to the 
attic reactor group. 

Fig. 5 shows the side of a re- 
actor unit without the tube panel, 
which plugs into the right-hand 
end. The tube panel hysterset 
assembly is shown just below the 
reactor. Mounted at the left end 
of the reactor is a contactor, one being supplied for every section 
controlled from the pilot board. 

Fig. 6 is a simplified diagram of the color master control, which 
operates only in conjunction with the rehearsal presets, indicated in 
the diagram as individual resistors 1, 2, and 3. The various color 
masters are connected to the same source of power as the other con- 
trol units. The individual rehearsal units may be transferred from 
independent bus to master control bus by small double-throw switches. 

Full-range control can be effected with the individual resistor con- 
trols provided the master variable autotransformer control is in the 
"full bright" position when the individual controls are thrown on the 
master bus; or, with the individual resistor controls when thrown on 




FIG. 5. ( Upper) A reactor unit, from 
the side, without the tube panel; 
(lower) tube panel hysterset assembly. 



228 



J. R. MANHEIMER AND T. H. JOSEPH [J. S. M. p. E. 



the independent bus. When on master control, the intensities of the 
various lamp groups can be varied collectively and proportionately by 
operating the master slider. 

Fig. 7 shows scene master and supplementary scene master for two 
scenes connected for fader operation. This simplified diagram does 
not show the interlocking selective buttons and switches for optional 
transfer to fader control. By operating the proper selective buttons, 
the fader can be preset and so connected that with a single operation 
of the fader lever one scene can be "faded out" and another scene 
"faded in" proportionately. After the fader lever has reached the 
limit of its motion, a new combination of the selective buttons can be 
chosen so that the existing scene may be "faded-out" and the next 
scene "faded-in," etc. 



MASTER. 



kUST^^S .NPEPTBUS 




FIG. 6. 



TO HYSTERSET3 

Simplified diagram of color master control. 



The lighting intensities of the various light sources correspond to 
the calibrated settings of the sliders on the individual presets, so that 
a combination of intensities from various light sources is possible 
ranging from nearly "black out" in certain groups to "full bright" in 
others. By setting some of the lights in a scene on a scene master and 
the remainder on the supplementary master, two master controls can 
be employed for each preset scene, permitting in effect ten preset 
scenes, although the board is known as a five preset board. Any 
set-up on the supplementary master can be transferred to the corre- 
sponding scene master without interrupting the continuity of the 
lighting by operating a small double-throw switch so as to put the 
entire lighting under the control of the one scene master. This fea- 
ture is required when it is necessary to "take out" in a single operation 



Mar., 1935] TUBE CONTROL FOR THEATER LIGHTING 



229 



all the lighting that may have been "brought in" by the operation of 
several controls. 

Fig. 8 is a photograph of one of the preset variable resistor control 
panel units. This unit measures 2*/4 inches in width and 12 inches in 
height ; which gives some idea of the compactness of this type of con- 
trol, considering that in this small space the equivalent of six dimming 
controls can be included. The space ordinarily occupied by one dim- 

SUPPLtMENTARY SCENE MASTER- 1 
SCEHE. MASTER 1. / M-2 SM-2 




TO RLHEABaAL 
CONTROL 



FIG. 7. 



Diagram of scene master and supplementary scene master, 
for two scenes connected for fader operation. 



ming plate and its corresponding circuit switch requires considerably 
more than this. 

At the top of this assembly is a pilot light, which indicates when 
the contactor in the reactor room closes. Below it are the preset 
controls for five preset scenes and one rehearsal. Each of the five 
preset scene controls is connected to a small double-throw switch used 
to transfer the corresponding preset from the scene master to the 
supplementary master, or vice versa. The rehearsal control is also 



230 



J. R. MANHEIMER AND T. H. JOSEPH [J. S. M. p. E. 



equipped with a similar small double-throw switch for transferring its 
control from the color master to the independent bus, or vice versa. 
Each of the quadrants of the individual presets is equipped with 

a calibrated scale to indicate the intensity 
of the light for which it is to be set. 
These calibrations are based upon visual 
brightness, and not upon photometrically 
measured intensities. 

This, in brief, describes the fundamental 
principles of the hysterset control. Of 
course, special features such as elaborate 
master control, preset control, extended 
control, etc., are attainable with this sys- 
tem. Its advantages are many compared 
with the commonly used systems. 

The tube is a simple two-element rec- 
tifier involving none of the delicate feat- 
ures of a grid-controlled tube. The opera- 
tion of the tube is not affected by changes 
in ambient temperature. It is possible to 
replace one tube with another without 
having to recalibrate the tube or the cir- 
cuit. The life of this type of tube is long, 
and the cost of replacement low. The 
equipment at the Center Theater has al- 
ready seen approximately 1500 hours of 
service without a tube failure and no in- 
dication of any. The current required to 
operate the controls is very small, a total 
of over 750 kw. of lamp load being con- 
trolled by less than 2 kw. at the control 
board. For example, the main ceiling 70 
kw. is controlled by a minute control con- 
suming but a few watts. Great flexibility 
of the control is achieved, and the change 
in light level is so smooth that a com- 
parable dimmer plate would need about 
750 contact buttons to equal it. As to compactness, the pilot 
board occupies a space 1 1 feet 5 inches long, 26 inches deep, and 6 feet 
6 inches high, and is operated by one man. Heat-producing appara- 




FIG. 8. One of the pre- 
set variable resistor control 
panel units. 



Mar . , 1935 ] TUBE CONTROL FOR THEATER LIGHTING 23 1 

tus is avoided on the stage, and the system is completely silent in 
operation. 

The real test of a switchboard is its effectiveness upon the audience 
in its control of light. The Great Waltz performance, now being 
staged at the Center Theater, demands the use of the entire 650 stage 
presets on the board, and light control changes occur approximately 
every ten seconds and at times every second. The equipment and 
entire system were installed under the supervision of the C. R. Place 
Engineering Associates. The switchboard was manufactured and 
supplied by the Westinghouse Electric & Mfg. Co., in combination 
with the hysterset control, which is a recent development of the 
Ward-Leonard Company. Similar equipment, but differing slightly 
in certain features, is at the present time being installed in the 
Metropolitan Opera House in New York, manufactured and supplied 
by the General Electric Company, and installed by the E-J Electric 
Installation Co. 

DISCUSSION 

MR. HASKELL: What is the difference between the board now installed at 
Rockefeller Center and other tube switchboards? 

MR. JOSEPH : The principle of most tube switchboards is the same, namely, to 
control a large current by means of compact and cool apparatus on the stage floor. 

Tube boards depend primarily upon the use of reactors for the actual dimming. 
The first reactor board for theaters was installed by the wiring department of the 
United Electric Light & Power Company in Daly's Theater at 29th Street and 
Broadway, in 1888. The change of reactance was achieved by pushing an iron 
core in and out. After the boardhad been completed, Mr. Daly was requested to 
observe its operation. The first core pushed in dimmed the lights, but hummed 
like a hive of angry bees. The next core was somewhat smaller, and the hum 
assumed a higher tone. With the third and fourth cores, the result was similar 
except for the tone. Mr. Daly looked and listened. He asked whether the noise 
was necessary, and was told that it was unavoidable. "Very well," he said, 
"throw the whole thing out into the street." That was the end of the first reactor 
board. 

Tube control can be broadly divided into two types : the Westinghouse-Ward 
Leonard type, in the Center Theater, and all the others. The control in the 
Center Theater, described in the paper, employs a regular full-wave rectifier tube, 
one for each reactor, for supplying the varying direct current required by the main 
reactor, and is controlled by a small reactor and associated apparatus called the 
hysterset. 

The other types furnish the varying direct current directly to the reactor 
through grid controlled tubes. For the larger reactors a second tube is added, 
and connected in parallel to the first. Other types use grid controlled tubes, 
which are cascaded, as in a radio receiver, progressing from small currents and 
small tubes to large currents and larger tubes. 



232 J. R. MANHEIMER AND T. H. JOSEPH 

Either a potentiometer or a small inductor may be used for varying the grid 
current at the pilot board. Moving the inductor armature increases or decreases 
the magnetic linkage and varies the pilot current accordingly. 

Each maker has different means of accomplishing the various master and fader 
controls. A complete comparison of these and other details would occupy more 
tune and space than is available. 

MR. HASKELL : How many tubes are used hi this board as compared with the 
other one? 

MR. JOSEPH: One tube for every circuit. On the previous board, which this 
replaces, eight tubes were hi the control unit and two tubes in the reactor set, 
making ten tubes per unit, instead of one. 



ROENTGEN CINEMATOGRAPHY* 
R. F. JAMES** 

Summary. This paper primarily treats of the advantages of x-ray motion pictures 
from the physician's and anatomist's points of view. To this end the author cites 
two examples in which a test apparatus making x-ray motion pictures has been used 
Jor diagnosis and for anatomical research. 

The requirements necessary for satisfactory x-ray motion pictures are outlined, 
and the limiting factors of the apparatus that has already been used are discussed. 
Descriptions of two test arrangements of apparatus are given. 

To the average individual, the thought of x-ray motion pictures 
brings a mental vision of skeletons, macaber-like, on the silvered 
screen. To the anatomist, and to the practicing physician, the 
thought of x-ray photographs capable of portraying the movements of 
the organs within the human body appears as a boon ; a veritable light 
in the wilderness. 

Far too few men realize the difficulties that beset the physician 
when he is called upon to diagnose the causes of human ailments. 
The diagnostician desires to locate the fundamental fault, and it is 
seldom that the patient is capable of describing accurately the 
symptoms or the conditions that preceded them. The physician is 
further handicapped by the fact that the human machine can not be 
easily and dispassionately disassembled for observation. 

Text-books are not entirely satisfactory, because they speak in 
general terms, and fail to consider the specific patient who has a 
confusion of symptoms. Successful diagnosis is largely made up of 
experience, sound judgment, and luck; all three factors are present in 
varying relations, and a stabilizing agent is urgently needed. 

Do not misunderstand the situation. There is ample proof that the 
effectiveness of the medical profession is steadily gaining, and a 
great measure of its gain is due to improved methods of diagnosis. 
The personal element is being assigned its proper relationship 
through the use of impersonal clinical and laboratory data. It is in 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 
** Research Dept., Westinghouse Lamp Co., Bloomfield, N. J. 

233 



234 R. F. JAMES [J. S. M. P. E. 

this particular field that the x-ray motion picture hopes to find its 
justification. 

As early as 1866 it was known that an abundant muscle tissue 
existed in the urinary transportation tract, but as late as 1931, a well- 
coordinated function of this muscle structure was denied. It was 
assumed that the kidneys filled up and naturally spilled over into the 
bladder. It appeared that the laws of gravity were universally 
applicable. 

Studies in pure anatomy had disclosed the muscular tissue but 
surgical observation of the moving organ was not conclusive. The 
motion could be attributed to a nervous spasm or to an involuntary 
response, and the time for a careful observation was necessarily 
limited by the conditions : the surgeon, in the press of an operation, 
has little time for contemplative meditation. 

The use of x-ray motion pictures, furnishing a progressive record 
of motion, has proved conclusively that the muscle tissue in the 
urinary tract has a functional purpose and that this function is well 
coordinated. It has also shown that disease will affect this co- 
ordinated function, and hence x-ray motion pictures furnish another 
diagnostic procedure. We know now that the normal kidney and 
the ureter pump their contents with a rhythmic motion and do not 
depend upon gravity. 

Unless a person has suffered from renal colic or other urinary 
disease this development does not hold much of interest. Another 
instance, more spectacular although of slightly less importance from a 
medical point of view, might be cited. 

Not long ago, in the community in which the research on x-ray 
motion pictures was being carried on, several new-born infants died. 
It was unfortunate that they died and still more unfortunate that they 
were all first children. It is hard to reconcile a couple who have just 
had the harrowing experience of losing their first baby. All too often 
they seek the doubtful solace of recrimination. 

The physicians attending the infants came to the conclusion that 
death was due to hypertrophy of the thymus, one of the ductless 
glands situated in the upper central section of the chest. There are 
cases where this gland grows to excess and causes death by strangula- 
tion. 

The death of the infants was attributed to the enlargement of 
this gland. Why it enlarged no one knew, and hence no one knew 
how to prevent the excessive growth. There must be some sort 



Mar., 1935] ROENTGEN CINEMATOGRAPHY 235 

of mental telepathy or "grape-vine telegraph" between expectant 
mothers, because the news of the "thymus deaths" spread rapidly 
despite the attempts at censorship, and soon a virtual panic was 
prevalent. The natural phenomenon of birth can not be conveniently 
postponed and the situation became critical. Deaths from hyper- 
trophy of the thymus are rare, but impersonal statistics give small 
comfort to distraught parents- to-be. 

At the height of the confusion it was decided to take x-ray photo- 
graphs of each new-born child. It was hoped that this procedure 
might allay the common fear. It worked out very nicely so far as 
the parents were concerned, but it threw the hospital staff into a 
panic because an alarming number of the photographs showed a large 
shadow in the chest in the general location of the thymus. For some 
peculiar reason none of the infants showed the typical symptoms of 
the disease, and subsequent pictures did not show the shadow. The 
phenomenon was so consistent that it was concluded that x-radiation 
in minute doses would reduce the thymus. In all cases where there 
was an apparent enlargement of the gland, the shadow was reduced 
by the amount of x-rays necessary to take only several pictures. 
This seemed to be an ideal opportunity to try the x-ray motion pic- 
ture, and to learn just how much energy was required for what ap- 
peared to be a specific treatment. 

An infant having a thymus that was photographically shown to be 
enlarged was photographed with the x-ray motion picture camera. 
The first exposure showed a normal chest, the second showed an 
enlarged thymus, the third showed the shadow descending in the 
chest, and the fourth exposure showed that the thymus shadow was 
gone it coincided with the shadow of the heart. 

The infant anatomy shortly after birth might be described as 
plastic, in that the organs have not permanently found their location. 
The peculiarity that was shown on the serial pictures was a peculiarity 
of all the photographs, and the entire situation was a case of mis- 
taken identity. The shadow attributed to the thymus was caused 
by the heart, and came and went with the pulse. In those cases 
that appeared to be hypertrophy of the thymus the x-ray "snap-shot" 
was taken when the heart was not in its customary position. 

These two instances are cited to illustrate the practical value of a 
means for providing a photographic record of functional motion. 
There is no question as to the value of the present x-ray plate, which 
provides a static record ; it is simply that the moving record provides 



236 R. F. JAMES [J. S. M. P. E. 

a greater scope of diagnostic security. With no disparaging thought 
it may be said that the present x-ray photograph holds a position 
analogous to that of the old stereopticon of forty years ago. 

It is quite true that present roentgenological technic has ways and 
means for observing motion. As a matter of fact, this technic was in 
general use prior to radiography. The fluorescent screen has been 
used since x-rays have been practicable, but it has certain definite 
limitations. Within the limits of present knowledge it appears 
that these limitations are inherent to the phenomenon of fluorescence. 

A fluoroscopic screen is simply a sheet of transparent material, 
such as glass, covered with a thin but adherent layer of certain 
combinations of chemical elements that possess the property of 
fluorescing or giving off light when exposed to the proper radiations. 
This screen is placed between the observer and the objective, in the 
path of the radiant energy. The variations in the density or absorb- 
ing power of the objective cause corresponding variations in the 
degree of fluorescence, with resulting shadow-pictures. 

The primary limitation of the fluoroscopic screen is that the screen 
does not provide appreciable light intensity. This fault contributes 
to the hazy image and poor definition. The fluorescent light may be 
increased by increasing the intensity of the primary radiation, but 
this is hazardous both for the patient and for the observer. Fluoro- 
scopic observation is similar to looking at faint silhouettes through a 
frosted glass. It is not clear vision with satisfactory contrasts, and, 
above all, it is not a record that can be preserved for careful study. 

The next logical step that was taken was the development of 
serial and periodic x-ray photographic plates. The period, or fre- 
quency, of exposure is timed so that the images form a record of 
successive positions, which may be translated into phases of motion. 
However, the number of images per unit of time required to analyze 
the recorded motion properly is a disputed point. The problem is 
different from that met in the ordinary forms of cinematography: 
it is not merely a question of obtaining an adequate number of 
images without subjecting the patient to a dangerous quantity of 
x-radiation. 

There is no question but that a "continuous" film, such as the 
present cinema with sixteen frames per second, would be desirable; 
but it is questionable as to whether or not it is necessary. With the 
exception of the heart, the remaining organs move with a relatively 
slow rhythm. There are doubtlessly anatomical regions of high 



Mar., 1935] ROENTGEN CINEMATOGRAPHY 237 

mobility, but the present x-ray technic does not provide sufficient 
contrast to render them visible. 

Two schools of thought exist as to the best manner of obtaining 
x-ray motion pictures. One school holds that the periodic image is 
satisfactory and safest. The other group advocates the use of the 
"continuous." Both schools have their special apparatus which, in 
turn, has its respective advantages and disadvantages. The prin- 
ciples of operation of each type of apparatus are relatively simple to 
discuss and, it might well be added, at that point the simplicity 
ceases. 

The apparatus for the "periodic" image consists essentially of a 
camera and an x-ray tube, with the object to be photographed 
placed between the two. The camera mechanically exposes and re- 
places the film, and thereby does away with the present manual 
technic. In the true sense of the word it can hardly be called a 
camera; nor, for that matter, can the resulting periodic images be 
called motion pictures. 

The camera is without a lens, for the obvious reason that a lens 
has no proper effect upon the radiations that act upon the film. 
The film itself is- a strip six to eight inches wide and about twenty feet 
long. It is fed from one spool to another as the exposures are made, 
the movement being actuated by a motor-drive controlled through an 
adjustable electric timer. The usual fluorescent intensifying screens 
are used to provide contrast to the picture. The camera is usually 
arranged to fit just below the top of the treatment table in place of the 
regular film holder, and it is brought into focus by moving the x-ray 
tube and observing the fluorescent image on the film intensifying 
screen. 

The rate of making exposures is optional with the operator and 
is based upon his experience, the selected rate being set on the 
timing device. The first section of film is moved into place from 
lead-shielded compartments in the camera, the intensifying screens 
are pressed against the film, and the timer closes the circuit applying 
the electrical energy to the x-ray tube for the desired time. 

After the exposure is made, a motor-driven mechanism releases 
the screens from contact with the film and moves the spools that bring 
a fresh supply of film into place, thus completing the cycle. As the 
film is moved it is in contact with an electrically grounded surface 
to discharge the accumulated static and prevent brush discharges. 

The resulting film is nothing more or less than a series of the 



238 R. F. JAMES [J. S. M. P. E. 

usual x-ray plates taken at intervals. Studies of these photographs 
may be made in several ways. The strip of film may be moved 
in front of an illuminated glass screen, which is usually adequate for 
preliminary study. For a more detailed consideration a different 
technic is used. The method is rather unique and entails con- 
siderable work, but the advantages justify the effort. 

Tracings are made from the strip of film showing the boundaries 
of the particular organ under study. These tracings are made on 
white paper and pasted onto a black background in proper order. 
The silhouettes are then photographed with a 16-mm. camera and 
the studies made from the 16-mm. film. The results are reminis- 
cent of the early animated cartoons, but they give a splendid im- 
pression of the movements and travel of the viscera. 




FIG. 1. Cut-out silhouettes of normal kidney. 

Fig. 1 shows the cut-out silhouettes of a normal kidney. There 
are no two images alike. If only one photograph had been taken, 
as is the usual custom, the diagnostic value of such a single photo- 
graph would have been doubtful. It is entirely conceivable that a 
diagnosis of disease might have been made, because there are indica- 
tions of pronounced kidney activity. These photographs were from a 
test case, and the activity was due to beer. 

Fig. 2 shows the conditions due to an obstruction in the low ureteral 
region. The over-distention is marked, and has caused a reversed 
peristalsis. The obstruction was removed by surgical means. 

There are objections to this method of studying the interior of 
the human anatomy. It takes time and infinite patience and is 
expensive. The original film is cumbersome and the cost prevents 



Mar., 1935] ROENTGEN CINEMATOGRAPHY 239 

many "retakes"; the tracing must be done by a skilled technician, 
and the final film has a pronounced "flicker." In spite of these 
faults it provides sufficient contrast so that physical measurements 
may be made. With care and a knowledge of the optics involved, 
it has been possible to make computations to determine the actual 
size of the organ and its periodic changes in volume. These computa- 
tions have been confirmed and found accurate within the experi- 
mental error. 

There is a second method of achieving the desired results which 
better merits the title of roentgen cinematography. This method is 
the photography of the moving image on the fluorescent screen, 




FIG. 2. Over-distention and reversed peristalsis, due to 
obstruction in low ureteral region. 

which is accomplished by synchronizing the shutter on the camera 
with the electrical impulses to the x-ray tube so that the object is not 
subjected to an overdose of radiation. The result more closely 
approaches the typical motion picture, although the number of frames 
per second is usually reduced because of the low intensity of visible 
light from the fluorescent screen. 

As a matter of fact, the limitations imposed by the fluorescent 
screen are the essential handicaps to this technic. The visible light 
given off from a fluorescent screen is not intense, and the images are 
not sharply defined. These conditions require that a wide-aperture 
lens be used with a high-sensitivity film and a relatively slow film- 
travel. 



240 R. F. JAMES 

These limitations are essentially of a mechanical nature and 
hence are being met and overcome. New developments are increas- 
ing the light from the fluorescent screens, faster films are becoming 
available, and new types of x-ray tubes that will withstand the rigors 
of rapid "on and off" service are on the market. The problem is one 
of removing the irksome "bugs"; the theory has been proved to be 
correct, and the usefulness is self-evident. 



MY PART IN THE DEVELOPMENT OF THE MOTION 
PICTURE PROJECTOR* 

THOMAS ARMAT** 



It is difficult to trace to its beginning and fix a date for the concep- 
tion of an idea that leads to an invention. Of the interesting impres- 
sions of my childhood, the one made by the toy known as the Zoe- 
trope was among the most outstanding. The idea that its principles 
might be applied to producing a series of consecutive instantaneous 
photographs of objects in motion, so as to reproduce the motion, was 
suggested by something I had read, and the fascinating thought per- 
sisted in my mind until the Anschutz tachyscope I saw at the Chicago 
World's Fair in 1893 brought a realization of its actual accomplish- 
ment. 

A toy magic-lantern was also one of my much prized playthings, and 
from it I had learned, among other things, that microscopically small 
objects could be projected upon a screen and greatly enlarged. My 
first thought upon seeing the tachyscope was of the possibilities that 
would be presented if its pictures could be projected upon a screen. 
The tachyscope was a peep-hole apparatus, and the picture I saw was 
that of an elephant trotting along in a most realistic manner. It 
was an outdoor scene, a foreign setting. The idea of bringing scenes 
from far and interesting countries and projecting them upon a screen 
before comfortably seated spectators, was an exciting thought. 

In the summer of 1894 I saw at Washington the first exhibition 
there of the Edison kinetoscope. It interested me greatly. About 
that time Mr. H. A. Tabb, who had known me since my boyhood days, 
and who also was a friend of both members of the firm of Raff and 
Gammon, exclusive agents for the kinetoscope, dropped into my 
office in Washington and endeavored to interest me in a business way 
in the kinetoscope. He gave me glowing accounts of the public in- 

* Prepared at the request of the Historical and Museum Committee; re- 
ceived Jan. 7, 1935. 
** Washington, D. C. 

241 



242 THOMAS ARMAT [j. s. M. p. E. 

terest in kinetoscope exhibitions and as to the profits to be made out 
of them. 

One of the places Mr. Tabb had in mind for the profitable exhibition 
of the kinetoscope was Atlanta, Georgia, anticipating the large 
crowds that would attend tfre Cotton States Exposition scheduled 
for the following year. 

After investigating the matter I told Mr. Tabb that I could not 
see anything very promising in the kinetoscope as a commercial proj- 
ect, but that I could see a lot in a machine of the kinetoscope type if 
the pictures could be projected upon a screen, and that I believed 
that I could devise such a machine. 

Mr Tabb's answer to that was that he did not believe it was pos- 
sible to project such pictures successfully, because he knew that Raff 
and Gammon had urged the Edison Company to produce such a ma- 
chine and that they had failed to do so, and he, therefore, did not be- 
lieve it could be done. From what I knew of stereopticons it did not 
seem to me that the problem presented insuperable difficulties, and 
I began a research to find out all I could as to the state of the art and 
what, if anything, had been accomplished in the way of projecting 
such pictures upon a screen, at the same time starting preparations for 
experimental work. 

I had been inventing for a number of years and had received several 
patents, among them No. 361,664, filed January, 1887, covering an 
automatic car-coupler, which had been developed to the point of 
making a model and which received some very favorable considera- 
tion. A subsequent patent, No. 521,562, filed March, 1893, covering 
a conduit electric railway system, also received favorable criticism 
from various sources, among others from Professor Louis D. Bliss, 
founder and head of the Bliss School of Electricity of Washington, 
D. C., who wrote me a letter in which he said, "It is most decidedly 
a model of perfection when compared with the crude system of 
the General Electric Co., and the cumbersome, complicated, and 
unreliable mechanism of the Wheeless system." 

In the fall of 1894 I enrolled as a student in the Bliss School, largely 
for the purpose of acquiring practical information as to handling an 
arc light that I proposed to use in my motion picture projection ex- 
periments. When I explained my purpose to Professor Bliss, he told 
me that there was another student in his school who was also inter- 
ested in motion picture experiments. A few days later, at one of the 
classes, Professor Bliss introduced to me C. F. Jenkins, the student 



Mar., 1935] DEVELOPMENT OF PROJECTOR 243 

in question. Jenkins was a stenographer in the Life Saving Service, 
a branch of the U. S. Treasury Department. 

It developed that Jenkins, with the cooperation and assistance of 
Professor Bliss and E. F. Murphy, the latter having charge of the 
Edison kinetoscope in the Columbia Phonograph parlors in Wash- 
ington, had assembled a modification of the Edison kinetoscope, in 
which all Edison parts, films, sprockets, etc., were used. Jenkins 
called this peep-hole machine a"phantoscope,"and applied for a patent 
on it November 24, 1894. The patent was issued as No. 536,539 
on March 26, 1895. As the patent shows, the Jenkins modification 
differed from the kinetoscope only in respect to the shutter. Instead 
of using a rotating shutter with a slit in it for exposing the continu- 
ously running film over a stationary electric light bulb, Jenkins ro- 
tated the bulb itself. This modification accomplished no improve- 
ment in results. It amounted to a somewhat different way of doing 
the same thing in a somewhat less efficient manner. Its only virtue 
consisted in the possible avoidance of certain claims in the Edison 
kinetoscope patent, in which a specifically described shutter was in- 
cluded as an element. These claims were cited by the Patent Office 
against the Jenkins application. 

Practically every night that we met at the Bliss School Jenkins 
urged me to join with him in experimental work to develop a motion 
picture projection machine. He was fully convinced that a success- 
ful projection machine could be built upon the principle of the con- 
tinuously running film of the Edison kinetoscope type of exhibiting 
machine. I was not so certain about that, but I felt that an experi- 
mental start had to be made and the sooner the better, and finally 
agreed on March 25, 1895, to join with Jenkins under an agreement 
which he prepared. In April or May of 1895 we completed a pro- 
jection machine built on the kinetoscope principle. The machine 
turned out to be a complete failure, for reasons now obvious to any- 
one familiar with motion picture projection problems. 

After that I took complete charge of further experimentation, at 
my own expense, and finally we produced the first projection machine 
ever made that embodied an intermittent movement with a long pe- 
riod of rest and illumination of the pictures on the film. Application 
for patent on this machine was filed on August 28, 1895, and later is- 
sued to Jenkins and Armat as patent No. 586,953 (Fig. 1). The pat- 
ent drawings were made from the machine itself, completed a short 
time before the application was filed. As may be seen, we mounted 



(Io Model.) 8 Sheets Sheet 2. 

C. F. JENKINS & T. ARMAT, 

>HANTOSCOPE. 
No. 586,953. Patented July 20, 1897. 






FIG. 1. Arrangement for providing a long period of rest and illumination, and 

quick shift of pictures. 



DEVELOPMENT OF PROJECTOR 245 

an Edison kinetoscope sprocket upon the mutilated gear illustrated 
in the patent. This arrangement gave the desired long period of rest 
and illumination and quick shift of the pictures, and demonstrated 
the value of the method. The machine, however, was a mechanical 
failure. 

The Edison films we used (the only kind obtainable at that date) 
were all taken at the rate of approximately forty per second. The 
machine could not run the films at more than half that speed, and it 
thus gave a slow-motion effect to all the scenes. It made a terrific 
noise. The sprocket and mutilated gear weighed more than a pound, 
and after a few experimental exhibitions the recesses in the driven 
gear were battered out of shape and made useless. The machine 
was never exhibited outside my office at No. 1313 F St., Washington, 
D. C. I still have the original sprocket and mutilated gear in my 
possession. 

Under date of August 30, 1895, Jenkins wrote his friend Murphy 
that the machine was a "grand success," but I regarded it as a com- 
plete failure so far as its having any commercial value was concerned, 
and addressed myself to the task of devising a practicable machine. 
This I accomplished a short time after the failure of the Jenkins and 
Armat machine, with a modification of the Demeny camera inter- 
mittent negative film movement, adapted to projection machine re- 
quirements. I hurriedly assembled a crude machine, tried it out and 
found it satisfactory. Immediately afterward I had a more sub- 
stantial machine made, and with it gave a number of successful ex- 
hibitions in my office to friends and acquaintances. An account of 
this machine was published in the Baltimore Sun of October 3, 1895. 

In the month of September, 1895, we took this machine to the Cot- 
ton States Exposition at Atlanta, Georgia. Subsequently I had two 
duplicate machines made and sent to us there for exploitation pur- 
poses. I obtained a concession from the Exposition authorities and 
built a theater in the grounds for giving exhibitions, with the thought 
that receipts from the theater would help to pay the exploitation ex- 
penses. The anticipated Exposition crowds did not materialize, the 
receipts were small, and a very considerable loss was incurred. 

While at Atlanta, Jenkins borrowed one of the three machines, 
saying that he would like to take it to Richmond, Indiana, to give 
some exhibitions to his friends on the occasion of his brother's wed- 
ding, and that he would be back in a few days. Jenkins gave an 
exhibition with this machine in his brother's store in Richmond, as 



246 THOMAS ARMAT [J. S. M. p. E. 

announced in the Richmond Daily Telegram of October 30, 1895.* 

After Jenkins' departure from Atlanta I made some important 
improvements in the machine, including a loop, or slack-forming 
means, that improved the exhibitions and greatly reduced the wear- 
ing of the films. Subsequently I remodeled the machine, to give it a 
more commercial form. 

In the month of December, 1895, I got in touch with Messrs. Raff 
and Gammon of New York, who were the exclusive agents for the 
Edison kinetoscope and films. My idea was to arrange for a supply 
of films. In reply to a letter to them asking that they come to Wash- 
ington to see my machine, I received an answer to the effect that they 
had no faith in motion picture projection machines, since they had 
endeavored to induce the Edison Company to produce one and they 
had failed to do so, and they did not believe motion pictures could be 
successfully projected. After a further exchange of letters Mr. Gam- 
mon agreed to come to Washington if I should pay his expenses, 
which I agreed to do. Mr. Gammon arrived with a sort of apologetic 
air of having been fooled into a wild-goose chase. When I took him 
into the basement of my office and threw a picture upon the screen, 
his attitude underwent a complete transformation. His excitement 
and interest were most apparent. 

The result of the interview was a contract under the terms of which 
Raff and Gammon undertook to furnish films and to manufacture a 
certain limited number of machines, and licenses were to be granted 
upon a royalty basis to users of the machines and films, with terri- 
torial restrictions. No machines under any circumstances were to be 
sold. The Edison Company was to make the machines from a model 
I was to send them. 

Mr. Edison wanted to see an exhibition of the machine before de- 
tails as to the number of machines to be made by him, the supply of 
films, etc., were to be decided. It was arranged that I should give 
Mr. Edison an exhibition. I sent a machine over to the Edison 
Works at Orange, N. J., and later Messrs Raff and Gammon and I 
went over from New York to give the exhibition. The exhibition 
took place in a large room in the Edison plant and the sheet was a 

*Editor's Note: It has been stated several times in the literature that C. F. 
Jenkins gave an exhibition with his projector at Richmond, Indiana, on June 6, 
1894, but no proof of this earlier date has been obtained by the Historical Com- 
mittee. A photographic copy of the Richmond Daily Telegram for Oct. 30, 1895, 
describing the showing on October 29, 1895, is in the files of the Committee. 



Mar., 1935] DEVELOPMENT OF PROJECTOR 247 



large one. Mr. Edison was obviously surprised at the excellence of 
the exhibition and so expressed himself. On the way back to New 
York Mr. Gammon told me that Mr. Edison had agreed to all our 
plans but expressed the opinion that we were planning to have more 
machines made than necessary. We planned to make eighty ma- 
chines at first, but Mr. Gammon said that Mr. Edison believed that 
fifty machines would be sufficient to cover the country. This (oft 
quoted) statement might seem strange coming from a man of Mr. 
Edison's vision, but it should be borne in mind that up to that date 
(February, 1896) no pictures of outside scenes had been taken by the 
Edison Company. The scenes were all such as had been taken in 
the Edison "Black Maria," as they called it, a sort of open-air, black- 
lined stage adapted to be rotated so as to face the sun. The necessity 
for bright sunlight was largely due to the high speed of taking. The 
pictures were restricted to such as could be taken in the limited space 
of the small stage, and they were all of vaudeville subjects. 

Arrangements were made by Raff and Gammon to introduce the 
machine, or rather its exhibitions, to the New York public, and I 
was asked to come to New York to supervise the installation and 
operation of the machine. This I did, and on the evening of April 
23, 1896, I gave at Koster and Bial's Music Hall in New York, the 
first exhibition ever given in a theater of motion pictures as we know 
them today, embodying, as such exhibitions do, the feature of rela- 
tively long periods of rest and illumination of each picture on the 
film. I personally operated the machine the first night. All the 
scenes shown, with one exception, were what might be called vaude- 
ville turns, or stage subjects. A crowded audience applauded each 
of the scenes with great enthusiasm. The one exception to the stage 
scenes was an outdoor scene that Raff and Gammon had succeeded 
in getting from Robert Paul, who by that date was experimenting 
with motion pictures in England. This scene was of storm-tossed 
waves breaking over a pier on the beach at Dover, England a scene 
that was totally unlike anything an audience had ever before seen in 
a theater. When it was thrown upon the screen the house went wild ; 
there were calls from all over the house for "Edison," "Edison," 
"speech," "speech." 

A graphic account of the exhibition was published in the New York 
Herald of May 3, 1896, and previously to that date, on April 4 the 
New York Journal and the New York World published long accounts 
of the exhibition that I had given at the Edison Works. 



No. 673.992. 



(He Model.) 



T. ARMAT. 
VITASCOPE. 

(Application fltod It& tt, 1800.) 



Patented May 14. 1901. 



3 Sbeeti-Sheet 2 




FIG. 2. Mechanism of the Vitascope. 



DEVELOPMENT OF PROJECTOR 249 

It should be here stated that, by mutual agreement, it was decided 
that Edison's name should be used in connection with the machine. 
This was done partly for the commercial advantage of the prestige of 
his name and partly because he was the producer of and had patents 
pending covering the films, an essential part of the machine, that he 
was to supply. Prior to this, when I had gotten the machine in all 
its details into what I considered practicable commercial shape, I 
applied for a patent on it on February 19, 1896 (Fig. 2), and selected 
Vitascope as a name for the machine. This name was applied to a pro- 
jection machine for the first time in this patent application, and it 
would seem that I added a word to the English language as the word 
Vitascope now appears in most modern dictionaries. The Vitascope, 
Edison Vitascope, so-called, made an immediate hit and was in great 
demand all over the country. 

Subsequently I invented and patented another projection machine 
with a greatly superior intermittent movement. This machine is 
shown in my patent No. 578,185 filed September 25, 1896, issued 
March 2, 1897 (Fig. 3). This intermittent movement is known as 
the "Star Wheel" or Geneva Cross movement, and it superseded all 
others by 1897 and is in use today in practically every motion picture 
theater the world over. It was not, however, a part of the Raff and 
Gammon arrangement, being a somewhat later development. The 
intermittent movement has been called, appropriately I think, the 
"heart" of the motion picture projection machine. In the early days 
this intermittent movement of my patent No. 578,185 was used in 
the Edison Projectorscope, the Powers Cameragraph, the Vitagraph, 
the Lubin machine, the Baird machine, the Simplex machine, and 
many other early machines. 

The Raff and Gammon licensing arrangement started off auspi- 
ciously and the financial returns were satisfactory, but troubles de- 
veloped shortly. None of my patents had been issued at that date, 
and the applications were still pending in the patent office, two of 
them involved in "interferences" which greatly delayed their issue. 
No patent protection could be given until patents were actually issued. 
Piratical machines began to appear, and, in the absence of patents, 
could not be stopped. Later on the Edison Company began to be 
slow in supplying films. Friction, for that reason among others, de- 
veloped between the Edison Company and Raff and Gammon. 
Still later the Edison Company began to market a machine that in- 
fringed my pending patents. As soon as my patents were issued I 



(Mo Model.) 

No 578,185 

e 



T. ARMAT. 

VITASOOPE, 

Patented Mar. 2, 1897 




*^^r 

FIG. 3. Intermittent movement employing the star-wheel, or Geneva cross. 



DEVELOPMENT OF PROJECTOR 251 

organized a company, to which I transferred my patents. Warnings 
were sent out to infringers, and suits were filed. In many cases the 
suits were rendered fruitless by the simple expedient of fading away 
on the part of the sued infringer. The Edison Company was mak- 
ing and selling large numbers of machines that they called projector- 
scopes which infringed no less than three of my patents. We notified 
users of the machines that they must promptly arrange to pay us 
royalties for their use or they would be sued for infringement and 
damages. The Edison Company notified users of projectorscopes 
they had sold that they would be protected against any suits that we 
might bring. That made it necessary for us to sue the Edison Com- 
pany. In the meantime a suit we had brought against the Biograph 
Company reached its final stage and was decided in our favor, and 
the company was enjoined. On the strength of that decision an in- 
junction was obtained against the Edison Company. The Edison 
Company had pending in the Patent Office an application covering 
the only successful method of taking motion pictures and an applica- 
tion covering the perforated film. 

So long as the Edison Company and my company were fighting 
each other, no exhibitor could give an exhibition without risk of 
being sued by one side or the other. I had pointed out a number of 
times to the Edison Company the obvious advantages of our getting 
together on some basis that would not involve the sale of projection 
machines, but without avail. After we obtained the injunction 
against the Edison Company they tried in various ways to obtain a 
license from my company under which they would be permitted to 
sell machines. To that I declined to agree. From the beginning I 
had refused to sell machines, or to license others to do so, for the rea- 
son that I felt that whatever monopoly we might be entitled to under 
our patents would be destroyed by any sale of machines; and I also 
felt that any profit we might make out of the sale of machines would not 
be remotely commensurate with the earning power of the machines 
themselves. I wanted a royalty from exhibitors, small enough not 
to be felt by them, but which in the aggregate would net a handsome 
income to my company. 

The suit against the Biograph Company was for an injunction and 
damages of $150,000. Damages were also asked in the suit against 
the Edison Company. Both companies posted bonds and prepared 
appeals. While damages in patent suits are rarely collectible, a 
favorable decision in an injunction suit where damages are claimed 



252 THOMAS ARMAT [j. S. M. p. E. 

creates a very uncomfortable feeling on the part of the defeated 
party and the holders of any of their securities. 

The American Mutoscope and Biograph Company had outstand- 
ing a bond issue of $200,000. Some of the bonds were held by the 
Empire Trust Company of New York, who took notice of the success 
of our suit for injunction and damages against the Biograph Company. 

Among the stockholders of the Empire Trust was J. J. Kennedy, a 
very distinguished consulting engineer as well as a man of rare busi- 
ness ability, who was requested by the Empire Trust Company to 
study the motion picture patent and commercial situation and work 
out a plan that would help the Biograph Company and their bond- 
holders out of their difficulties. Mr. Kennedy got in touch with 
Mr. H. N. Marvin, also an engineer of distinction and an inventor, 
who was the president and general manager of the Mutoscope and 
Biograph Company. Together Mr. Kennedy and Mr. Marvin, 
after holding consultations with all interested parties, formed a stock 
company to take over all the valuable patents in the art, the stock 
to be distributed to the patent owners. It was a closed corporation, 
the stock was placed in escrow, and none of it was sold. 

This holding company was called the Motion Picture Patents Co., 
and the principal beneficiaries were the Edison Company, the Bio- 
graph company, and the Armat Moving Pictures Company. I 
owned most of the stock in the latter. The Motion Pictures Patents 
Company was an immediate success. The royalties that it collected 
put no burden upon the industry but resulted in a large net revenue 
to the Patents Company. A royalty of half a cent a foot was paid 
by the producers, and a royalty of two dollars a week was paid by the 
exhibitors to the Patents Company. 

At the date of the organization of the Patents Company there were 
in this country between ten and twelve thousand small theaters, or 
Nickleodeons, as they were called. The royalty of two dollars a week 
was an entirely negligible sum to them, but, as it was collected without 
cost to the Patents Company by the simple expedient of having the 
distributors add two dollars a week to their weekly film rentals, it 
amounted to a practically net revenue of between $20,000 and 
$24,000 a week. The revenue of half a cent a foot as film royalties 
also amounted to a handsome total. Unfortunately for the stock- 
holders of the Patents Company, as the Motion Picture Patents Com- 
pany came to be known, its life was rather a short one. 

Some of the producers, for reasons that I have never quite under- 



Mar., 1935] DEVELOPMENT OF PROJECTOR 253 

stood, were refused licenses by the Patents Company. These pro- 
ducers, calling themselves "Independents," formed an organization 
and put up an all-around fight. At that date anything that smacked 
of being a monopoly or trust was very unpopular with the public and 
the courts. 

The Independents charged the Patents Company with being an 
unlawful monopoly under the Sherman Anti-Trust law, and insti- 
gated a suit by the Government against them on that ground. In a 
decision by Judge Dickinson it was held in substance, as I recall it, 
that while a patentee had a legitimate monopoly within his patent 
claims, he could not, under the Sherman act, lawfully combine his 
patent with other patents, and the Patents Company was ordered 
dissolved. 

I have always felt that Judge Dickinson was influenced in his judg- 
ment by the fact that the Edison Company (under the domination 
of Gilmore) had sold thousands of projection machines without re- 
strictions as to their use, in some instances guaranteeing the right to 
their use, and, later, through the Patents Company, participated in 
royalties collected for their use. 

Judge Dickinson said, "Every theater was required to pay royal- 
ties for the use of projection machines, even where the machine had 
been owned before the combination was formed." He appeared to 
overlook, or to ignore, the fact that the machines had been sold with- 
out license or other authority from the owners of the projection ma- 
chine patents. 

I have always felt that the Patents Company, instead of being an 
organization in restraint of trade, the thing that the Sherman law 
was designed to prohibit, was in effect an organization to facilitate 
trade; for the reason that prior to the date of the Patents Company's 
acquiring the right to grant licenses, under all the controlling patents, 
no producer or exhibitor could do a legitimate business that is, a 
business that did not infringe one or more patents and the fear of 
running counter to the patent laws could certainly have had a deter- 
rent effect upon the business of all except those piratically inclined. 

Many erroneous statements have been made and published as to 
when and by whom the first motion picture projection machine was 
made. To clarify the facts I have been asked several times to list the 
more or less basic inventions upon which the motion picture industry 
was initially established, as shown by U. S. Patent Office records. 
I have been regarded as qualified to do so because of my own pioneer 



254 THOMAS ARMAT [j. S. M. P. E. 

inventions in the art and my connection with the beginning of the 
industry founded upon them. Subsequently to this early experience 
I was called upon to testify, as an expert in the art, in litigation under 
my patents, and later under the Edison and other patents owned by 
the Patents Company. 

There have been a great variety of motion picture projectors, pro- 
duced under different names, that vary as to their mechanical details 
but embody all the inventions that may be called basic basic in the 
sense that they are necessary for successful projection and have been 
used since the beginning or near the beginning and are still being used. 
The following is my list of the eight most important inventions in 
the motion picture art : 

(1) The Edison camera: Patent No. 589,168, dated Aug. 31, 1897. Filed 
Aug. 24, 1891. This was the first camera employing a perforated film which was 
given an intermittent motion so that a given number of perforations and a given 
number of pictures would be intermittently moved, rather than a given length of 
film. The result was a film having equally spaced, juxtaposed pictures through- 
out its length. The first practicable motion picture camera ever produced. 

(2) The Edison motion picture film: Patent Reissue No. 12,038, Sept. 30, 1902. 
Filed Aug. 24, 1891. The first perforated motion picture film ever produced hav- 
ing equally spaced, juxtaposed pictures, necessary to successful motion picture 
projection and an essential part of every motion picture projector in use the 
world over today. This Edison film when first made some time prior to 1891 was 
! 3 /8 inch wide over all, contained four perforations to each picture, the picture 
itself being 1 inch wide by 3 /4 inch high. The number of perforations per 
picture and the film dimensions have not been changed in standard size machines 
since they were first made by Edison some time prior to 1891. 

(3) The Edison peep-hole kinetoscope: Patent 493,426, dated March 14, 1893. 
Filed Aug. 24, 1891. This was the first motion picture exhibiting machine em- 
ploying a perforated film with equally spaced, juxtaposed pictures. The first 
practicable motion picture exhibiting machine of any kind, but incapable of project- 
ing pictures successfully because it gave the film a continuous motion instead of 
an intermittent motion. 

(4) The Jenkins and Armat intermittent motion projection machine: Patent No. 
586,953, dated July 20, 1897. Filed Aug. 28, 1895. The first motion picture 
projection machine giving the pictures an intermittent motion with a long period 
of rest and exposure. A mechanical failure, it nevertheless demonstrated the 
necessity and value of long exposure, essential to successful projection. 

(5) The Vitascope: Invented and patented by Thomas Armat, Patent No. 
673,992, dated May 14, 1901. Filed February 19, 1896. The first projection 
machine employing a loop-forming means and the first projection machine em- 
bodying a practicable intermittent movement giving the pictures the required 
long period of rest and exposure. A loop-forming means is essential in projection 
machines employing a long length of film. 

(6) The star-wheel intermittent movement: Invented and patented by 



Mar., 1935} DEVELOPMENT OF PROJECTOR 255 

Thomas Armat. Patent No. 578,185, dated March 2, 1897. Filed September 
25, 1896. By means of this intermittent movement a small sprocket carrying the 
film could be given a gradually accelerated intermittent movement without film 
wear and tear and without jar to the mechanism. This movement superseded 
all others by 1897, and has been continuously used up to date. The intermittent 
movement is called the "heart" of the projecting machine. 

(7) The Albert E. Smith framing device: Patent 673,329, dated April 30, 
1901. Filed March 15, 1900. This device frames the pictures while the machine 
is running, and is a practically essential device. 

(8) The John A. Pross shutter: Patent 722,382, dated March 10, 1903. Filed 
January 19, 1903. An important improvement for reducing scintillation or flicker. 
Not so essential in the earlier days of 1895 and 1896 when Edison films were the 
only ones obtainable, since these films were taken at approximately forty per 
second, but quite essential with pictures taken at the later commercial lower rates. 

The foregoing is a complete list of the pioneer inventions covering 
all the essentials of the motion picture camera, the motion picture 
film, and the motion picture projector, and they are all in universal 
use today in the most modern and up-to-date equipment. The addi- 
tion of color and of sound accompaniment belong to a later period. 

For the possible benefit of those who have not investigated the 
matter, I believe it might be well to point out some of the differences 
between a camera and a projection machine, from the patent and in- 
vention standpoint. These differences were pointed out by me in 
the Patent Office interference in which my Vitascope patent, No. 5 
on the list, was involved. I am not an attorney, but my familiarity 
with the art and its requirements enabled me to conduct this case 
successfully myself, preparing the brief and arguing the case person- 
ally before the several tribunals of the Patent Office and the Court of 
Appeals of the District of Columbia, all of which tribunals accepted 
my views and decided in my favor. In taking a picture of an object 
in motion it is essential to make the exposure of the image on the 
.sensitive film as short as possible, consistent with the sensitiveness 
of the film, for the reason that if this is not done there will be time for 
the image of the moving object to be displaced on the sensitive film, 
causing a blurred or indistinct picture. In an exhibiting apparatus 
the reverse is true. There is, in the exhibiting apparatus, a picture 
fixed beyond the possibility of any such image movement's causing 
blur, and the longer the picture is exposed to the eye, the better the 
results. In a camera we are dealing with a moving object and a sen- 
sitive film. In an exhibiting apparatus we are dealing with a fixed 
picture and the human eye. No question of flicker or scintillation 



256 THOMAS ARMAT 

enters into the problem of taking pictures. That question enters 
very extensively into the problem of exhibiting pictures. 

In a camera, the sensitive film does not cooperate with the mecha- 
nism to produce a complete or final result. The film has to be taken 
out and developed and printed before the operation is complete. 
The film is run through the camera but once. The Patent Office and 
the Courts held that the film is no more a part of a camera than the 
paper is of a printing press. In an exhibition machine the film with 
pictures on it is an essential part of the apparatus. It is a part of the 
mechanism which cooperates with the other parts to produce the 
complete and final results. In an exhibition machine the film is used 
over and over again in the apparatus and has to be so used whenever 
the apparatus is used. In passing upon this question the Patent 
Office had this to say: 

"If Latham with his Exhibit Machine No. 12, and Casler with his Exhibit First 
Machine, both of which were taking cameras, could, without invention, have pro- 
duced a machine of the construction called for by the issue, it is remarkable that 
they did not do so at any proven date before the filing of their application. 

The evidence shows that neither Latham nor Casler was an ordinary mechanic 
but that they were inventors of considerable capacity, and yet neither of them 
produced a machine having the new and beneficial results which are claimed for 
the machine described in Armat's application." 

The Patent Office said further: 

"In our opinion, proof of the existence of a camera for taking pictures of 
object in motion, said camera having in combination with a sensitive film, me- 
chanism for giving the film an intermittent motion in which the periods of pause 
exceed the period of motion, said mechanism comprising in addition the other 
elements called for by the issue and a shutter, is not a reduction to practice of this 
issue ; unless there is proof to show that when this camera was used for projecting 
the shutter was either omitted altogether, or was so adjusted as to provide for 
such relative periods of pause and illumination and periods of motion as are 
called for by the issue." 






OVERCOMING LIMITATIONS TO LEARNING WITH THE 
SOUND MOTION PICTURE* 

V. C. ARNSPIGER** 



Summary. Although many limitations to learning have been overcome by the 
inventions of such tools as the telescope and the microscope, education has had, to a 
large extent, to depend upon the printed word. The introduction of the sound mo- 
tion picture, in numerous forms and combinations, makes available for education 
the benefits of all such tools, including the advantages of slow-motion, time-lapse, and 
animated cinematography, and sound recording and amplifying devices. The edu- 
cator is thus enabled to transcend many obstacles in the way of presenting and clarify- 
ing abstract concepts. 

There are certain limitations to human learning which have 
seriously restricted the development of school curricula from the 
kindergarten to the post-graduate level of instruction. Outstanding 
of these limitations are: the difficulty with which the individual 
acquires concepts which depend almost wholly upon verbalism for 
their presentation; his inability to perceive certain movements in 
nature because of the rapidity or slowness with which they occur; 
his inability to see objects which, because of their extremely small 
size or because of then- great distance in space, are beyond the range 
of the unaided human eye; his inability to hear sounds which, be- 
cause of their extremely small volume or because of their great dis- 
tance from the hearer, are beyond the limits of unaided perception; 
and his inability to reach backward into the past and reproduce 
objects and actions which contribute to the conditioning of his 
present environment. 

These rather obvious limitations have to some extent been over- 
come by mechanical means which have been developed in the course 
of scientific research during the nineteenth and twentieth centuries. 
The modern telescope has enabled the individual to make excursions 
into space which until recently were impossible. The microscope 
has brought into the range of human vision objects which, without 

* Presented at the Spring, 1934, Meeting at Atlantic City, N. J. 
** Erpi Picture Consultants, New York, N. Y. 

267 



258 V. C. ARNSPIGER [j. s. M. p. E. 

its assistance, would have been forever beyond perception. 
The motion picture camera has extended the arm of the interpretive 
artist so that for a good many years he has been able to record, for 
the eyes of the future, objects in action which hitherto would have 
passed into unreality to be reproduced only by means of indefinite 
forms of verbalism. The slow-motion and time-lapse devices 
which have been added to this instrument have contributed to the 
clarification of many concepts which in the past were beyond the 
perception of the individual learner. Modern sound recording, ampli- 
fying, and transmitting devices have made many significant and 
fundamental contributions upon which the reality of concepts de- 
pends. 

Although these products of modern invention have been available, 
some of them, for many years, education has to a great degree con- 
tinued to depend largely upon the printed word for the transfer of 
ideas. This dependence upon the product of the printing press has 
resulted in many restrictions in the scope and range of the curriculum. 
The aims and objectives of education have been established some- 
times with, and often without, a conscious realization of these re- 
strictions imposed by this mechanical invention. A glance into the 
activities of the traditional primary school will reveal the nature 
of these restrictions. At the time of his entrance into the school, 
the normal pupil has during the first five or six years of his life al- 
ready attained a rather profound mastery of several tools of learning 
in the form of the five senses. Through the use of these tools, learn- 
ing from his immediate environment has proceeded at a very rapid 
rate. By the time he starts to school, however, the returns from his 
immediate environment have begun to diminish. In order for full 
development to continue, a significant elaboration of this environ- 
ment must be furnished. This elaboration of environment is one 
of the responsibilities which the school must assume. It is generally 
true that this institution is so organized that it is literally forced to 
await the mastery of additional tools of learning before it can pro- 
ceed at a very rapid rate toward the fulfillment of this responsibility. 
It is needless to recall the emphasis which at present is placed upon 
these "tool subjects" reading, writing, and arithmetic of the tradi- 
tional elementary school curriculum. By far the most important of 
these is reading. Many concepts which are fundamental to the 
future elaboration of the pupil's environment he can not acquire with 
the optimal degree of mastery through the use of this "tool" because of 



Mar., 1935] LEARNING WITH SOUND PICTURES 259 

the complexity of certain mental processes which learning through 
reading presupposes. 

The sound picture has at its disposal, in numerous desirable com- 
binations, the inherent advantages of the telescope, the microscope, 
the motion picture camera (with its many devices such as slow- 
motion, time-lapse photography, and animation), and sound record- 
ing and amplifying devices. Clarification of concepts through this 
medium is not necessarily dependent upon the "tool subjects" of the 
curriculum. Thus, the sound picture holds out to those respon- 
sible for the construction of the curriculum a unique challenge. 

This challenge can be met by the application of instruments 
of research, many of which are now available, some of which have yet 
to be developed. What effect, if any, will the use of these mechani- 
cal devices have upon the mastery of the time-honored "tool subjects" 
of the present curriculum? Are there objectives of education which 
are withheld until the later years of school life solely because of this 
dependence upon the mastery of concepts which are too "difficult" 
to present through reading? Can certain objectives be attained 
with the aid of these products of modern invention? Can much time 
of the school be saved and energy of teachers conserved for more im- 
portant duties? 

It is possible that the answers to these questions would release the 
curriculum specialist from many of the restrictive influences which 
are inherent in the present methods of subject matter presentation, 
to the end that the extension of the curriculum would be deter- 
mined not by the present limitations to learning, but by the needs of 
the race. 

During the past few years films have been produced which were 
designed to transcend the obstacles to learning in a manner not pos- 
sible by means of traditional methods of instruction. The educa- 
tional sound-film Fundamentals of Acoustics demonstrates these facts 
more impressively than would be possible by talking about the 
problem in an abstract way. In this film abstract phenomena are 
visualized intangible phenomena that can not be perceived by means 
of the visual senses; such as, for example, the movement of sound- 
waves through the air. By means of animation it is possible to repre- 
sent the sound-waves visually and in motion, with their characteris- 
tics under various conditions. In the traditional methods of in- 
struction, such phenomena were generally taught by reading and 
lectures. The student's comprehension of the concepts involved de- 



260 V. C. ARNSPIGER 

pended to a very large degree upon his comprehension of the printed 
page. In reducing these abstract and difficult concepts to a sound- 
film, the student is enabled to comprehend them irrespective of hi< 
reading ability. His progress is then determined solely by his in- 
nate mental ability. 

During the past few decades, the older subjects have become fairl} 
well established at certain grade levels in the school curriculum. Foi 
example, physics is usually first taught in the junior or senior year oi 
the high-school. This is largely due to the fact that it was supposed 
and probably rightly, with prevailing methods of instruction, thai 
this subject could not be properly comprehended at a lower level 
But through a medium such as the sound-film, in which abstract anc 
intangible concepts can be made, in effect, concrete, it is now possible 
to present them at a much lower level, possibly even in the advancec 
grades of the elementary school. 

The film Fundamentals of Acoustics reproduces certain sound ef 
fects and demonstrates the alteration of speech and music by limit 
ing the frequency range or by attenuation and change in the qualit} 
of sound. Demonstrations such as these are difficult, and in mos 
high-schools impossible, due to limitations of equipment and thi 
tremendous cost that would be involved referring particularly t< 
such scenes as those dealing with reverberation in different kind 
of rooms; the limitation of the frequency range ; echoes; andoutdoo 
attenuation. Attention is also called to the possibility of visualizinj 
invisible phenomena, which is well illustrated in this film by ai 
authentic sequence on the functioning of the human ear. Finally 
one of the most important contributions resulting from the use o 
such films in school instruction should be emphasized the number o 
concepts that can be presented in the mere span of ten minutes. Aj 
a result, the concepts become closely integrated, and the relationship 
of one to the other become self-evident. 



ROLLER DEVELOPING RACK FOR CONTINUOUSLY 
MOVING THE FILM DURING PROCESSING BY 
THE RACK-AND-TANK SYSTEM* 

C. E. IVES** 



Summary. In order to obtain machine type development in the rack process, a 
ack has been devised by means of which the film is moved continuously over rollers 
luring processing. The film traces a helical path passing over rollers carried by an 
ipper and a lower shaft, and is then led back along the bottom of the rack where a 
iosed loop is completed by splicing the ends together. 

The film is propelled by rotation of the upper shaft and rollers which, in turn, are 
Iriven by a motor and a reduction gear. The lower shaft is allowed a slight parallel 
lisplacement in the vertical direction to compensate for changes in the length of the 
llm which occur when it is wetted. 

A rewind and roll holder are mounted upon a removal plate for use in loading and 
mloadingfilm or leader, and the film is removed to a reel for drying. A wide variety 
>/ processing conditions have been met successfully in processing sound and picture 
tegatives and prints. 

Continuous machines now in almost universal use for processing 
motion picture film in the larger laboratories are, for many reasons, 
poorly adapted to experimental work. The time of treatment 
allotted for any stage of the process can usually be varied only slightly, 
while the processing solutions which are ordinarily supplied from a 
recirculating system containing a large volume of liquid can not be 
:hanged readily. Consequently, the execution of an experimental 
procedure requiring the use of new processing solutions and any con- 
siderable variation in the time of treatment is slow and extremely 
costly. 

When developing by the rack-and-tank method, 1 the degree of 
uniformity of development is not satisfactory for precision sound 
recording owing to the formation of rack marks, 2 but it was con- 
sidered that if the film could be moved continuously, a degree of 
uniformity could be attained comparable to that attainable with com- 



* Presented at the Fall, 1934, Meeting at New York, N. Y. Communication 
No. 539 from the Kodak Research Laboratories. 
** Eastman Kodak Co., Rochester, N. Y. 

261 



262 C. E. IVES [J. S. M. p. B 

mercial processing machines. Accordingly, a roller rack was con 
structed which was capable of moving a 200-foot length of film con 
tinuously, as in the usual developing machine, but which was port 
able and could be used in conjunction with the tanks of the type use< 
in the rack developing process. 




FIG. 1. The roller developing rack threaded with film. 

Although a rack capable of handling film in the manner describee 
must have parts which function similarly to the corresponding parti 
of a continuous developing machine, the literature reveals no attempts 
to use such a device for this purpose. A rack described by Crab 
tree 1 - 2 was equipped with rollers to permit a slow movement of tht 



Mar., 1935] A ROLLER DEVELOPING RACK 263 

ilm backward and forward through a distance of a few feet for 
the purpose of eliminating the sharply defined rack markings which 
Otherwise occurred where the film rested against the upper and lower 
cross bars, but the type of treatment afforded by continuous rapid 
motion of the film as in a developing machine was not attained. 

The Film Path. In studying possible paths for the film on the rack, 
; the arrangement commonly used in the rack-and-tank process, as well 
is in developing machines, was chosen. The film is carried in a spiral 
Dr helical path over a succession of rollers at the top and bottom of the 
rack, the upper and lower rollers each having a common shaft. In 
order that a 200-foot length of film may be allowed to move con- 
tinuously in one direction, it must form a closed loop when threaded 
on the rack. This requires a path for returning the film to the start- 
ing point after it has traversed the rack. The arrangement shown 
in Figs. 1 and 2 by which the return path is located along the 
bottom of the rack, was finally chosen. Thus, starting at one end, 
the film reaches the other end of the rack by following a helical 
path, turning around rollers along the top and bottom, leaves the 
last upper roller to go to the lower corner of the rack while making 
a quarter turn, traverses the length of the rack on the supporting 
rollers along the bottom, and then after making another quarter 
turn arrives at the starting point. 

Since the film increases in length slightly when it is wetted, provi- 
sion was made for automatically increasing the length of the loops 
to the extent of l*/4 inches. Various materials, as for example, 
coated film and uncoated leader, expand in various degrees, and for 
this reason it is not sufficient that the lower shaft be free to rise and 
fall, but it must do so in such a manner as to redistribute the slack 
created locally by this differential expansion. This is accomplished by 
compelling the lower shaft to remain parallel to the upper shaft while 
it moves up and down. If parallelism is maintained, then any slack 
which appears while the film is running is immediately redistributed, 
because the shorter strands receive the full weight of the lower shaft. 
If, on the contrary, the lower shaft were allowed to tip, any slack or 
excess tension appearing at one end would be accommodated by 
tilting of the shaft, whereupon some of the remaining loops would 
hang clear of the roller flanges and become overlapped. 

One of the simplest ways of maintaining the lower shaft parallel 
to the upper in its vertical motion is probably that sometimes em- 
ployed in developing machines. It consists in including the lower 



264 C. E. IVES [J. S. M. P. E. 

shaft in an approximately square rigid frame which slides along the 
side members of the rack structure, as on rails. This type of guide 
is not so well suited to a rack which is long in the horizontal direction 
because it requires large heavy parts to maintain the necessary 
rigidity in the guiding members. Therefore, this scheme was dis- 
carded in favor of a device which relies on two evener plates con- 
nected to the ends of the lower shaft assembly by short rigid links and 
to each other by Vie-inch stainless steel* wires. Other methods of 
effecting this parallel motion are dealt with later. 

In order to support the paralleling mechanism and guide the 
lower shaft in its vertical motion, a supporting structure was provided 
at both lower corners of the rack, and space was left for the vertical 
return path of the film between the side members of this struc- 
ture. 

The Motor Drire. Although a large stationary drive would be more 
advantageous for use with a number of racks, it was decided to mount 
an individual drive on this experimental unit. The motor ( 1 /2oth-hp.) 
and reduction gear (1 :10) are contained in a stainless steel box 
at one end of the rack, above the upper shaft. The repulsion- 
induction motor running at 1725 rpm. drives the reduction gear 
through a F-belt running over pulleys chosen to give a linear speed of 
film travel of 85 feet per minute. The upper shaft is driven from the 
reduction gear by a Vs-inch, ] /2-inch pitch stainless steel roller chain 
and 7-tooth stainless steel sprockets (1 to 1). Some advantage 
might be gained by the use of a suitable motor with built-in re- 
duction gear, but materials available on the market at the time were 
used. 

The housing was used to shield the drive from corrosive liquids 
and the film from oil and corrosion products. Since the housing 
hinders the dissipation of heat from the motor, connections are pro- 
vided so that compressed air can be used for cooling after long runs. 
Inasmuch as the upper rollers are completely immersed during proc- 
essing, the chain carries along appreciable quantities of the processing 
solutions, and in order to protect the interior of the drive housing, the 
reduction-gear shaft was supplied with a splashproof fitting where 
it passed through the end wall. As shown in Fig. 2, this consists 

* The expression "stainless steel" is used to signify alloy steels containing 
chromium and nickel in the proportion of, roughly, 18 per cent chromium and 8 
per cent nickel, and sometimes described as "super-stainless steels." This is the 
only metal used for parts of the rack proper. 



Mar., 1935] A ROLLER DEVELOPING RACK 265 

of a flanged extension cap of stainless steel which, with the sprocket 
at the outside, makes a splashproof closure. 

The Rack Frame. Although the drive elements are far from over- 
size for the work, their weight, in addition to that of the other neces- 
sary parts, is sufficient to cause the rack frame to bend somewhat 
during handling. A frame stiff enough to resist the force of this 
weight would be too heavy to be handled readily. 

With this realization, the uprights and upper beam (Fig. 2) were 
made of wood, and the bracket, corner gussets, and the channel 
at the bottom were of stainless steel to save weight, where practicable, 
and still maintain a reasonable degree of rigidity. Attachments at 
the top for suspending the rack in the processing tanks allowed it 
to hang free of strain during processing. 

Shaft Bearing and Roller Assembly. In order to avoid misalign- 
ment of the upper roller shaft and the wiper shaft in their respective 
bearings, self -aligning bearings were adopted by the use of free-fitting 
pins and a spring spider, as shown in the detail of Fig. 2. The bear- 
ing at the middle of the shaft is given a large clearance. 

This problem does not arise in connection with the lower shaft, 
which is held rigidly in line by its own beam with which it floats 
practically free of the rack frame, as regards distortion. 

Although the film is driven by virtue of the rotation of the upper 
shaft and rollers, a certain amount of differential motion among both 
upper and lower rollers is necessary to assist in the redistribution of 
slack when a difference in length exists in the strands at different parts 
of the rack. This freedom to turn is allowed to two out of every three 
rollers. The type of roller shown in Fig. 2 has only one flange for 
economy of space. On the upper shaft the spacing is regulated by 
fastening every third roller by means of a stainless steel set-screw; 
the spacing requires checking occasionally. It will be noticed that 
the rollers are separated into two groups by the middle rack bearing, 
at which point a single flange is added to each group to complete 
the last roller. The shaft is secured in position with respect to 
the axial direction by a collar at one end and the drive sprocket at 
the other, both of which rest against the end shaft bearings. 

Some difficulty was expected at points where a stainless steel shaft 
ran in stainless steel bearings without lubricant, although at the 
slow speed of operation and with ample bearing clearances it was 
not believed that the bearings would seize. When operated dry, 
even at moderate speeds, the bearings were found to be scored and it 



266 



C. E. IVES 



[J. S. M. P. E. 



was necessary to refinish them. At this point the practice of 
lubricating with a small drop of liquid soap whenever bearings were 
run dry was commenced. This dry condition was found only when 
the rack had not been in use for processing for several hours. No 
further trouble has since appeared. Recently it has been found that 




FIG. 2. Assembly drawing of roller rack. 

synthetic resin compositions gave satisfaction when inserted as bush- 
ings in similar service. 

The lower roller shaft is maintained rigidly in line by the use of a 
stainless steel beam which at the same time supplies, by its weight 
(15 pounds), the necessary tension on the loops. While this weight is 
undesirably large, it is necessary in order to overcome friction in the 
evener and to maintain sufficient tension while the rack is being 
lowered into the liquid. All rollers on this shaft are free to turn, the 
shaft remaining stationary at all times. In order to prevent exces- 



Mar., 1935] A ROLLER DEVELOPING RACK 267 

sive spacing between the rollers, a spring was placed at either end of 
the shaft. This spring is rather short and is shaped in such a way that 
only a very light force is applied when the roller stack-up clearance is 
taken out, but a much greater force as soon as the rollers move apart. 
Thus, under normal conditions, frictional resistance where the rollers 
come together is very slight. These special provisions for mainte- 
nance of exact spacing are necessary only in the case of rollers which 
have a flange at one side only. With such rollers, excessive spacing 
permits the film to run out of position to such an extent that the 
roller treads are in contact with the portion of the film opposite the 
picture or sound-track. This danger is obviated by the use of 
rollers having double flanges which, however, require appreciably 
more space. If the lower shaft were not kept parallel to the upper 
by the evener, this simple means would be inadequate for maintaining 
proper spacing. 

The Evener. The action of the evener is as follows : Referring to 
the drawing (Fig. 2) and the close-up view (Fig. 3), it will be seen that 
the extremities of the lower shaft assembly are connected to the 
triangular evener plates by links, the evener plates being mounted so 
that they rotate about a pin carried by side members from the frame. 
On each plate and equidistant from the center of gyration are attached 
swivels to which are fastened crossed tension wires from the other 
plate. 

If force is applied to raise the right-hand end of the shaft as- 
sembly, the link is put under pressure, causing the right-hand plate to 
rotate in a counter-clockwise direction. This results in tension in 
the wire attached at the bottom of this plate which, in turn, causes 
tension to be applied at the top of the plate on the opposite end of the 
rack, thereby producing clockwise rotation, tension in the link at that 
end, and elevation of the shaft assembly. The other possibilities are 
obvious. 

Lower Shaft Assembly Guide, In order to make the film loops as 
long as possible, the lower shaft is located below its beam, with the 
result that the assembly is somewhat top-heavy. In order to main- 
tain correct vertical alignment, the end of the beam is held in a clip 
lying between the evener supporting members (see detail in Fig. 2) 
which are shaped in such a way as to form guides at this point. 
In addition, a prolongation of the roller shaft at either end is fitted 
with a collar which travels between guides (Fig. 3) . 

Vertical motion is limited by means of the connecting link wrist- 



268 



C. E. IVES 



[J. S. M. P. E. 



pin which is extended at both sides to pass through slots in the side 
plates. Swinging of the assembly in the direction parallel to the 
shaft axis is also prevented by this pin. Clearances at all these 
points are such as to prevent any interference with the necessary 
movement of this assembly by a distortion of the frame which might 
occur in normal use. 

Method of Use. The rack is prepared for use by completely thread- 
ing the film path with leader, which also remains upon the rack when 
it is not in use. When film is to be loaded, a plate bearing the feed 
stock and rewind is mounted on the rack, as shown in Fig. 4. The 
end of the film to be processed is attached to one end of the leader 




FIG. 3. Close-up of evener mechanism. 



by means of the mechanical splicer, while the other end of the leader 
is led to the rewind. The motor is started and leader is run off as the 
film is led onto the rack. During the loading operation a tension is 
maintained at the rewind and feed roll so that the lower shaft 
assembly is kept near the upper limit of its range. When all the film 
is transferred (215 feet max. length), the leader is cut off at the rewind 
and spliced to the end of the film. The motor is run again for a 
moment to redistribute any slack. 

The rack is then placed in the developer by two men who grasp 
it by the upright ends (Fig. 5). The motor is started as soon as 
the film is fully immersed, and is not stopped until the instant the 



Mar., 1935] 



A ROLLER DEVELOPING RACK 



rack is to be removed from the developer. Immediately after the 
rack is first placed in the developer, the sheet-rubber wiping pads are 
brought down to the position shown in Fig. 6, where they sweep off 
any airbells which may have become attached to the film during 
immersion of the rack. At the end 
of one-half minute the wiper shaft 
is rotated by means of the small 
crank, bringing the wipers to the 
position shown in Fig. 2, where it 
is locked by the spring catch visible 
below the crank. When the end 
of the development time is reached, 
the motor is stopped and the rack is 
transferred to the fixing bath, fol- 
lowing a definitely timed sequence 
of movements. The film is kept in 
motion during fixation and during 
part of the washing operation. 

Processing operations are timed 
by means of an electrically driven 
darkroom clock. The duration of 
any treatment may be as long as 
desired, but a development time of 
less than 1 minute requires a degree 
of precision in timing the transfer 
operation which is difficult to 
attain. 

Running speeds of 40 and 85 feet 
per minute have been attained by 
changing two pulleys in the drive. 

Drying. The film is transferred 
to a reel for drying in the manner 
illustrated in Fig. 7, loose water 
being removed by means of the 
pneumatic squeegee. 3 This opera- 
tion is somewhat different from the 

loading process. Inasmuch as the leaving strand is not under tension, 
it is found desirable to diminish resistance on the incoming strand by 
unthreading the return path at the bottom of the rack so that leader 
is drawn directly to the last lower roller as shown. At this time the 




FIG. 4. View of the demountable 
feed stock and rewind as used in 
threading the rack. 



270 C. E. IVES [J. S. M. P. E. 

small soft-rubber guide roller is brought down to the position shown 
in Fig. 6 to increase the traction on the film drive roller. The 
frame supporting the guide roller rotates freely about the wiper shaft 
between two collars. It is held either in the operating position or 
vertically upward by the flat spring shown in Fig. 6. 

Before using the rack again, loose water is removed by the applica- 
tion of a blast of compressed air to the rollers. 

Uniformity of Development. The rack was found capable of produc- 
ing very uniform development when operated at 85 feet per minute. 
In order to attain the highest degree of uniformity, a compressed-air 
agitator was placed in the developing tank during processing. This 
produced a worthwhile improvement in uniformity but caused ex- 




FIG. 5. The rack in operating position in the developing tank. 

cessive scratching of the film base by making it run against the sta- 
tionary parts of the rack within the film loops. Various remedies 
were tried. One consisted in stretching sheets of dental-dam rubber 
over these parts but was unsuccessful because the wet film and 
rubber adhered so strongly when the rack was out in the air that 
the rack could not be unloaded. A solution was found in the use 
of Kodatape as a cover for all these surfaces, except for a small 
area where application of the rubber was more convenient and 
caused no interference. 

A desirable form of compressed-air distributing device consists of 
astainless steel pipe resting upon the bottom of the tank, drilled 
to provide outlets at regular intervals. In order to attain good 
distribution of air or other gas it is necessary to make holes just 
large enough to deliver the correct volume of air at the working 
pressure. If the holes are too large the distribution is uneven, and if 
too small, they are easily obstructed. A better method of securing 
good distribution of air at varying pressures is to cover the pipe 



Mar., 1935] 



A ROLLER DEVELOPING RACK 



271 



with a loosely fitting rubber tubing fastened at either end by a wire 
clamp and perforated with nail holes at frequent intervals. Since 
the holes in the rubber adjust themselves to the flow of air and do 
not become clogged with sediment, it is thus possible to use large 
holes in the pipe. 

The uniformity of density of a flashed film processed on the roller 
rack was found to be equal to that obtained by the best available con- 
tinuous machine processing. 

A technic has been devised for processing the necessary H&D 
strips for a time-gamma determination in a single run, and then in a 
subsequent run obtaining a check 
of any desired point on the time- 
gamma curve. 

Table I shows typical data 
chosen at random from the re- 
sults of individual runs made 
with a number of different emul- 
sions and developers. It is seen 
that the gammas obtained were, 
for the most part, within 3 per 
cent of the required values. In 
one case where the developing 
time was only 1 minute and 25 
seconds, the error was greater 
than this, which is an indication 
of the difficulty of accurately 
timing the manipulation of the 
rack in transferring from the de- 
veloper to the fixing bath. The wide variety of conditions met 
may be judged by the fact that the gammas ranged from 0.52 to 3.08 
with times of development of 1 minute 25 seconds to 15 minutes. 
The degree of accuracy indicated is likewise attained in duplicate 
runs. 

The rack has been used successfully for both picture and sound 
records with various emulsions and developers. 

Replenishing. A replenishing solution described previously 4 was 
used successfully for maintaining the activity of the developer. It 
was added at the rate of one-half gallon of double-strength replenisher 
to a 75-gallon tank for every 8 minutes of development with air 
agitation. The rate of replenishing was apparently governed largely 




FIG. 6. Close-up of upper shaft 
wiping device, and guide roller. 



272 C. E. IVES [J. S. M. P. E. 

by the action of the air used in agitating the developer, since it was 
based successfully upon the time of treatment instead of upon the 
footage developed. 

Alternative Methods of Obtaining Parallel Motion. A number of 
alternative mechanisms for producing a parallel motion of the lower 
shaft assembly have been considered as follows : 

(a) It is evident by reference to Fig. 8(B) that if the evener plates 
were cut in the form of circles so as to function as pulleys for the 




FIG. 7. Method of transferring the film to the drying reel. 

wire, more latitude would exist in the choice of proportions for the 
plates. 

(b) Fig. 8(C) illustrates a hydraulic evener consisting of metallic 
bellows at either end of the lower shaft with cross-connected piping. 
Such an arrangement would eliminate frictional resistance almost 
entirely. Tension in the loops could be maintained by the use of an 
additional bellows at the center of the beam connected with the 
water or compressed air line, thus eliminating the need for much of 
the weight in the beam. A valve in the pipe leading to this bellows 
could be used conveniently to lock the lower shaft assembly in posi- 
tion while the rack was being moved from one tank to another, 



Mar., 1935] A ROLLER DEVELOPING RACK 273 

TABLE I 

Data from Individual Runs Following the Indications of Gamma-Time Curves 

Time of Development 
Gamma Required from Curve Gamma Obtained 

0.52 9Min. 15 Sec. 0.52 

0.60 2Min. 50 Sec. 0.60 

0.60 2Min. 12 Sec. 0.61 

0.60 2Min. 35 Sec. 0.60 

0.60 2Min. 33 Sec. 0.58 

0.60 2Min. 36 Sec. 0.60 

0.60 2Min. 6 Sec. 0.60 

0.65 4Min. 30 Sec. 0.64 

0.65 8Min. 15 Sec. 0.64 

1.3 2Min. 10 Sec. 1.27 

1.5 2Min. 40 Sec. 1.51 

1.5 2Min. 30 Sec. 1.42 

1.5 2Min. 30 Sec. 1.47 

1.5 6Min. 30 Sec. 1.56 

1.7 4Min. 1.74 

1.8 IMin. 25 Sec. 1.94 
1.8 4Min. 1.90 
2.0 4Min. 2.05 
2.0 4Min. 1.98 
2.0 4Min. 2.01 
2.0 4Min. 2.00 
2.0 4Min. 1.97 
2.0 3Min. 48 Sec. 2.00 

2.2 IMin. 50 Sec. 2.26 

2.3 15Min. 2.27 

2.4 5Min. 2.45 

2.4 15Min. 2.49 

2.5 8Min. 30 Sec. 2.49 

2.6 5Min. 30 Sec. 2.65 
3.0 5Min. 3.00 
3.0 8|Min. 3.08 

(c) A device similar to that used on drafting boards, consisting of 
loops of wire or cable passing around pulleys and cross-connected to 
the two ends of the shaft assembly to be maintained in alignment. 

(d) Other devices working on the rack-and-pinion principle 
were considered. In one arrangement the racks are secured to the 
developing rack frame and are shackled to small pinion gears on a 
shaft carried by the lower roller assembly. In another application 
of this principle, the long pinion gear shaft under torsion was elimi- 
nated by mounting small segments of large gears rotating on pins set 
in the beam associated with the lower roller shaft. These gear seg- 




(AS EMPLOYED) 




FIG. 8. Schematic drawing of evener mechanisms. 



ments engage each other at the midpoint of the beam and mesh with 
the racks located at the ends of the main developing rack frame. 

Acknowledgment. Grateful acknowledgment is made of the assis- 
tance and suggestions given by Messrs. S. Forsyth and W. Bahler 
during the construction and testing of this apparatus. 



REFERENCES 

1 CRABTREE, J. I.: "The Development of Motion Picture Film by the Reel 
and Tank Systems," Trans. Soc. Mot. Pict. Eng. (May, 1923), No. 16, p. 163. 

2 CRABTREE, J. I., AND IVES, C. E.: "Rack Marks and Airbell Markings on 
Motion Picture Film," Trans. Soc. Mot. Pict. Eng. (Oct., 1925), No. 24, p. 95. 

3 CRABTREE, J. I., AND IVES, C. E.: "A Pneumatic Film Squeegee," Trans. 
Soc. Mot. Pict. Eng., XI (Aug., 1927), No. 30, p. 270. 

4 CRABTREE, J. I., AND IVES, C. E.: "A Replenishing Solution for a Motion 
Picture Positive Film Developer," /. Soc. Mot. Pict. Eng., XV (Nov., 1930), No. 5, 
P. 627. 



WILLIAM VAN DOREN KELLEY 

. 

(1876-1934) 

William Van Doren Kelley was born at Trenton, N. J., March 8, 
1876. While a youngster in the Trenton schools, he exhibited a 
talent for invention and in addition, became interested in theatricals. 
Thus was laid the foundation for his later activities in the field of the 
motion picture. His ability as an inventor and his interest in the 
theater led him to stage acts of magic and illusion with the acumen 
and showmanship of a much older person. 

While still young, "Bill" began to earn his living as a show-card and 
window-display artist. This activity led to a partnership in an adver- 
tising sign business with a brother, George Kelley, in Brooklyn, N. Y. 

The history of their business was marked by the invention of 
many successful devices, among them being the "flashing light" 
type of sign introduced about 1910, which became very popular 
and made a good income for the company. 

In the meantime, as funds became available from the profits of 
the business, Kelley devoted his spare time to working on a process 
for coloring motion pictures. He had first become interested in 
the problem after a trip with Joseph Mason to England for Biograph 
in 1910. Kelley 's job at that time was to decorate and make atten- 
tion-attracting devices for the Biograph showhouses in England, and 
he also worked with the Biograph camera. He realized the value that 
color would give to the films, and from then on his cherished dream 
was to invent a process for coloring motion pictures. Every spare 
cent and every moment not used for earning a livelihood was directed 
toward achieving this goal. 

In 1913 he formed a company called "Panchromotion" for the 
development of an additive color process somewhat similar to the 
Kinemacolor of that time. In order to minimize color fringing, 
one of the deficiencies of such systems, he increased the number of 
pictures taken in a given unit of time. He also tried to improve the 
color rendition by using ihree and four colors in the color- wheel on the 
projector; the Kinemacolor used only two 

275 



276 WILLIAM VAN DOREN KELLEY [j. S. M. P. E. 

The color experiments were conducted in the basement of a house 
at 1586 E. Seventeenth St., Brooklyn, N. Y. During this time a 
double-coated stock and a bleach formula which had much to do 
with the success of the later Prizma process were perfected. From 
Brooklyn the Panchr emotion Company moved to quarters in a vacant 
garage in Jersey City, N. J. By this time, a certain measure of 
success had attended Kelley's endeavors, and Prizma Incorporated 
was formed with sufficient capital to undertake regular production. 

Subsequently to 1916, Prizma sent cameramen with the Prizma 
filter-wheel cameras throughout the world to make travel and nature 
pictures. The negative films were returned and finished at the 
Prizma laboratory. 

The first Prizma film was Our Navy, released in 1917 at the Forty- 
Fourth Street Theater in New York City, and also shown about the 
same time at the Strand Theater in that city. . The color was pro- 
duced by an additive process, using a color-wheel on the projector. 

Kelley was not satisfied, however; he believed that the color could 
be applied directly to the film by a subtr active system. In order to 
carry out this idea, he entered a partnership with Carroll H. Dunning 
and Wilson Saulsbury, and a laboratory was opened at 205 W. 
Fortieth Street in New York City under the name "Kesdacolor." 

Their first film made by the subtractive process was a picture of 
the American flag. In a length of fifty feet it was shown at the Roxy 
and Rialto at New York, on September 12, 1918. 

Shortly after the success of this showing, Kelley returned to the 
Prizma Company, which was reorganized. Longer films were under- 
taken, and in 1919 a single-reel travel subject was subtractively 
colored. 

J. Stuart Blackton of Vitagraph saw this picture and was so 
impressed that he decided to make a feature-length picture in 
Prizma color. The Glorious Adventure was released in April, 1922. 

Kelley continued his researches far afield in a search for yet un- 
tried methods in color. In 1919 he produced a series of colored 
animated cartoons which were drawn by Pinto Colvig; in 1923 he 
developed a stereoscopic motion picture novelty in color; in 1924 he 
introduced Kelley-Color which was an imbibition process. In this 
last-named system, two colors were imbibed upon a black-and-white 
key image. In 1926 he became associated with Max Handscheigl in 
the formation of the Kelley-Color Company which was bought by 
Harris-Color in 1928. In 1929 Kelley started his experiments with 



Mar., 1935] WILLIAM VAN DOREN KELLEY 277 

the bi-pack negative method, which has since been widely used in 
making "process shots" and color-separation negatives. 

During the last days of his life, Kelley was working upon a system 
in which the bi-pack negative films were cemented together before 
perforating, thus assuring better registration and a closer union of 
the two strips of film during the photographing. The cemented 
films were to remain cemented until after development. 




W. V. D. KELLEY 

The only medal ever issued by the Society of Motion Picture Engi- 
neers was presented to him on October 13, 1919, "for achievement in 
color motion pictures." 

Kelley contributed many papers dealing with color photography 
to the Transactions of the Society, and served as chairman of the 
Color Committee for many years. 

The Society and the industry sustained a great loss in Bill's passing. 

W. E. THEISBN 



SPRING, 1935, CONVENTION 
SOCIETY OF MOTION PICTURE ENGINEERS 

HOTEL ROOSEVELT, HOLLYWOOD, CALIF. 
MAY 20-24, INCL. 

Officers and Committees in Charge 

PROGRAM AND FACILITIES 

W. C. KUNZMANN, Convention Vice-President 
J. I. CRABTREE, Editorial Vice-P resident 
J. O. BAKER, Chairman, Papers Committee 

LOCAL ARRANGEMENTS AND RECEPTION COMMITTEE 

P. MOLE, Chairman 

G. S. MITCHELL E. HUSE G. F. RACKETT 

W. QUINLAN K. F. MORGAN H. W. MOYSE 

J. A. BALL W. C. HARCUS J. A. DUBRAY 

C. W. HANDLEY E. C. RICHARDSON F. E. JAMES 

R. H. McCuLLouGH R. G. LINDERMAN C. DREHER 

PROJECTION COMMITTEE 

H. GRIFFIN, Chairman 

]. O. AALBERG R. H. MCCULLOUGH 

L. E. CLARK K. F. MORGAN 

Officers and Members of Los Angeles Local 150, I. A. T. S. E. 

STUDIO AND NEW EQUIPMENT EXHIBIT 

O. F NEU, Chairman 

H. GRIFFIN J. FRANK, JR. 

P. MOLE S. HARRIS 

BANQUET 

W. C. KUNZMANN, Chairman 

P. MOLE W. QUINLAN E. HUSE 

G. S. MITCHELL G. F. RACKETT S. HARRIS 

PUBLICITY COMMITTEE 

W. WHITMORE, Chairman 
J. J. FINN A. JONES 

F. H. RICHARDSON P. A. McGuiRE 

G. E. MATTHEWS 
278 



SPRING, 1935, CONVENTION 279 

MEMBERSHIP 

O. M. GLUNT, Financial V ice-President 
E. R. GEIB, Chairman, Membership Committee 

LADIES' RECEPTION COMMITTEE 
MRS. E. HUSE, Hostess 

assisted by 

MRS. G. F. RACKETT MRS. F. E. JAMES MRS. E. C. RICHARDSON 

MRS. W. QUINLAN MRS. P. MOLB MRS. F. C. COATES 

MRS. C. W. HANDLEY 

Headquarters 

The headquarters of the Convention will be the Hotel Roosevelt, where excellent 
accommodations and Convention facilities are assured. Registration will begin 
at 9 A.M. Monday, May 20. A special suite will be provided for the ladies at- 
tending the convention. Rates for S. M. P. E. delegates, European plan, will be as 
follows : 

Single : $2.50 per day ; one person, single bed. 

Double: $3.50 per day; two persons, double bed. 

Double: $4.50 per day; two persons, twin beds. 

Suites: $6.00 and $8.00 per day. 

Technical Sessions 

An attractive program of technical papers and presentations is being arranged 
by the Papers Committee, laying special emphasis upon the developments in the 
technic, equipment, and practices of the studios. Several sessions will be held in 
the evening, to permit those to attend who would be otherwise engaged in the 
daytime. All sessions will be held at the Hotel. 

Studio and Equipment Exhibit 

The exhibit at this Convention will feature apparatus and equipment developed 
in the studios, in addition to the usual commercial equipment. All studios are 
urged to participate by exhibiting any particular equipment or devices they may 
have constructed or devised to suit their individual problems, conform to their 
particular operating conditions, or to achieve economies in production, facilitate 
their work, or improve their products. 

Those desiring to participate should communicate with the General Office of 
the Society, Hotel Pennsylvania, New York, N. Y. No charge will be made for 
space. Each exhibitor should display a card carrying the name of the particular 
studio or manufacturer, and each piece of equipment should be plainly labeled. 
In addition, an expert should be in attendance who is capable of explaining the 
technical features of the exhibit to the Convention delegates. 

Semi- Annual Banquet 

The semi-annual banquet of the Society will be held at the Hotel on Wednesday, 
May 22. Addresses will be delivered by prominent members of the industry, 
followed by dancing and entertainment. Tables reserved for 8, 10, or 12 persons; 
tickets obtainable at the registration desk. 



280 SPRING, 1935, CONVENTION 

Studio Visits 

S. M. P. E. delegates to the Convention have been courteously granted the privi- 
lege of visiting and inspecting the Warner Bros. First National Studio, the Fox 
Hill Studio of Fox Film Corp., and the Walt Disney Studio: admission by regis- 
tration card only. A visit has also been arranged to the California Institute of 
Technology. 

Motion Pictures 

Passes will be available during the Convention to those registering, to Grau- 
man's Chinese and Egyptian Theaters, Pantages', Hollywood Theater, Warner 
Bros.' Hollywood Theater, and Gore Bros.' Iris Theater. 

Ladies 9 Program 

An especially attractive program for the ladies attending the Convention is 
being arranged by Mrs. E. Huse, hostess, and her Ladies' Committee. A suite 
will be provided in the Hotel where the ladies will register and meet for the various 
events upon their program. 

Further details of the Convention will be published in the next issue of the 
JOURNAL. 



SOCIETY ANNOUNCEMENTS 

PACIFIC COAST SECTION 

The regular monthly meeting of the Section, held on January 31 at the Los 
Angeles Museum, was attended by almost a hundred members and guests. 
Through the courtesy of the Museum, not only the Motion Picture Exhibit, 
but also adjacent sections were opened to the members. Much favorable com- 
ment was expressed concerning the Motion Picture Exhibit and the efforts made 
by Mr. W. E. Theisen, Honorary Curator of the Exhibit and chairman of the 
Historical and Museum Committee of the S. M. P. E., in building it up. 

Opening the meeting, Mr. G. F. Rackett, Chairman of the Section, first out- 
lined the history and aims of the Society for the benefit of the guests, and then 
introduced Mr. Theisen, who presented an excellent talk on the history of motion 
pictures, beginning with references to the persistence of vision by the ancient 
writers. Carrying the subject through the many and varied devices developed 
throughout the years, the history terminated with the description of Edison's 
application of the motion picture film produced by George Eastman. 

Following Mr. Theisen's talk, several pictures were shown illustrating the vast 
improvements in the art of depicting motion, and covering the span from 1898 
to the present time. Included were three reels of Fairbank's Robin Hood, of 1921, 
and one of the most recent Technicolor "Silly Symphonies" The Tortoise and the 
Hare, by courtesy of Walt Disney. 

PROJECTION PRACTICE COMMITTEE 

At a meeting held at the Paramount Building, New York, N. Y., January 23, 
the topics to be included in the Spring report of the Committee were discussed. 
In addition to other subjects, it is planned to include a revision of the report of 
August, 1931, containing the recommended arrangements for projection rooms, 
bringing it up to date and amending it wherever necessary hi the light of recent 
developments and the experience of the past few years. 

STANDARDS REPRINTS 

Reprints of the Standards Adopted by the S. M. P. E. and Recommended 
Practice are now available in booklet form. They may be obtained from the 
General Office of the Society, at the price of twenty-five cents each. 



The Society regrets to announce the death of one of its members: 

JAMNADAS SUBEDAR 

October 9, 1934 



281 



282 SOCIETY ANNOUNCEMENTS 

SOCIETY SUPPLIES 

Certificates of Membership may be obtained from the General Office by all 
members for the price of one dollar. Lapel buttons of the Society's insignia are 
also available at the same price. 

Black fabrikoid binders, lettered in gold, designed to hold a year's supply of the 
JOURNAL, may be obtained from the General office for two dollars each. The 
purchaser's name and the volume number may be lettered in gold upon the back- 
bone of the binder at an additional charge of fifty cents each. 

Requests for any of these supplies should be directed to the General Office of 
the Society at the Hotel Pennsylvania, New York, N. Y., accompanied by the 
appropriate remittance. 



JOURNAL 



OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 

Volume XXIV APRIL, 1935 Number 4 



CONTENTS 

Page 

Some Photographic Aspects of Sound Recording 

C. E. K. MEES 285 

Certain Phases of Studio Lighting C. S. WOODSIDE 327 

Historical Notes on X-Ray Cinematography 

R. F. MITCHELL AND L. G. COLE 333 

Symposium on Construction Materials for Motion Picture 

Equipment : 
Applications of Stainless Steels in the Motion Picture Industry 

W. M. MITCHELL 346 

Laminated Bakelite in the Motion Picture Industry 

R. L. FOOTE 354 

Inconel as a Material for Photographic Film Processing Appa- 
ratus F. L. LAQUE 357 

Spring, 1935, Convention at Hollywood, Calif 372 

Society Announcements 377 



JOURNAL 

OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 



SYLVAN HARRIS, EDITOR 

Board of Editors 
J. I. CRABTRBE, Chairman 

O. M. GLUNT A. C. HARDY L. A. JONES 

J. O. BAKER 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Ijic., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1935, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Sjociety of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 



Officers of the Society 

President: HOMER G. TASKER, 4139 38th St., Long Island City, N. Y. 
Past-President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
Executive Vice-President: EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, 

Calif. 

Engineering Vice-P resident: LOYD A. JONES, Kodak Park, Rochester, N. Y. 
Editorial Vice-President: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Financial Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Convention Vice-President: WILLIAM C. KUNZMANN, Box 6087, Cleveland, Ohio. 
Secretary: JOHN H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York, N. Y. 

Governors 

MAX C. BATSEL, Front & Market Sts., Camden, N. J. 
LAWRENCE W. DAVEE, 250 W. 57th St., New York, N. Y. 
ARTHUR S. DICKINSON, 28 W. 44th St., New York, N. Y. 
HERBERT GRIFFIN, 90 Gold St., New York, N. Y. 
WILBUR B. RAYTON, 635 St. Paul St., Rochester, N. Y. 
SIDNEY K. WOLF, 250 W. 57th St., New York, N. Y. 



SOME PHOTOGRAPHIC ASPECTS 
OF SOUND RECORDING* 

C. E. K. MEES** 

Summary Since 1928, a very considerable amount of study has been devoted 
to the photographic problems which arise in connection with recording and reproducing 
sound. A general understanding has been reached as to the principles involved and 
as to the conditions which will permit satisfactory quality in the sound reproduced. 
This lecture represents a summary, therefore, of the general conclusions which have 
been reached by those interested in the subject. 

Sound has three attributes, namely: (1) loudness, (2) frequency or pitch, and (5) 
wave-form, quality, or timbre. For perfect reproduction of sound, the reproduction 
should be perfect in respect of all these three attributes. The intensity range is limited 
primarily by the ground noise, and secondarily by the modulation which is permissible 
before the wave-form becomes sufficiently distorted to be noticeable to the ear. The 
ground noise is due primarily to physical defects in the films, such as scratches and 
dirt, although even in' a perfectly clean film there is a very small amount of ground 
noise due to the granular structure of the silver deposit. This ground noise can be 
diminished by systems of noise reduction. 

The reproduction of high frequencies is dependent upon the resolving power of the 
photographic film, and the effect of loss of these high frequencies will be illustrated. 
Special apparatus has been designed for an investigation of the quality of the reproduc- 
tion, including special sensitometers and densitometers. The analysis of the quality 
of reproduction and the application of graphic analysis are demonstrated, with an 
illustration of some of the results attained. 

During the last five or six years, the technical problems relating to 
the production and exhibition of motion pictures have been changed 
very greatly by the introduction of the recording and reproduction 
of sound. The introduction of sound recording has, indeed, in- 
fluenced every part of the motion picture industry, from the nature 
of the original material selected for presentation to the architectural 
design of the motion picture theater. Any attempt to consider 
within the confines of a single paper the whole of the changes re- 
sulting from that introduction would necessarily lead to a very super- 
ficial result, but a brief and general discussion of the photographic 

* Presented at the Fall, 1934, Meeting at New York, N. Y. Communication 
No. 530 from the Kodak Research Laboratories. 
** Eastman Kodak Co . Rochester, N. Y. 

285 



286 



C. E. K. MEES 



[J. S. M. P. E. 



problems which arise in connection with the recording and reproduc- 
tion of sound may have some interest and value for the general scien- 
tific public. 

Since 1928, a very considerable amount of study has been devoted 
to these photographic problems, and a general understanding has 
been reached as to the principles involved and the conditions which 
produce satisfactory quality in the sound reproduced. This paper 
represents a summary, therefore, of the general conclusions which 
have been reached by those interested in the subject, attention be- 
ing directed especially to the technical methods of sound recording 
which have been developed in the United States. 

Many of the investigations which are 
summarized here have been carried out 
by my colleagues in the Kodak Research 
Laboratories, and my thanks are due to 
them and especially to Dr. Otto Sandvik 
for their assistance in the preparation of 
this paper. 

The sound record on .a film is contained 
in the sound-track, a narrow strip occupy- 
ing an area between the picture area and 
the perforations of the film, as is seen in 
Fig. 1. This sound-track contains a pho- 
tographic record so prepared that when it 
passes through a very narrow beam of 
light during the projection of the picture, 
the intensity of the light transmitted by 
it varies rapidly and continuously in the 
same sense as the air-pressure represent- 
ing the sound by which the record was produced. By means of a 
photoelectric cell, the transmitted light is converted into electrical 
energy which, after amplification, actuates a loud speaker and re- 
produces the sound. The cycle of operations concerned in the re- 
cording and reproduction of sound therefore may be visualized some- 
what as is shown in Fig. 2. 

The desire to make a simultaneous record of sounds and actions 
is not new; it was, indeed, the objective toward which Edison strove 
in his early work on both the cinematograph and the phonograph. 
But its practical realization depended upon the discovery of some 
method of increasing the amount of energy corresponding to a given 




FIG. 1. Variable-density 
sound-track on film. 



April, 1935] 



SOUND RECORDING 



287 



sound, and this in turn depended upon the interchangeable conversion 
of sound energy into electrical energy and upon the possibility 
presented by the electrical valve tube of increasing or amplifying 
electrical currents. Thus, in Fig. 2, a source of sound, /, emits pres- 
sure waves which the microphone, II, converts into electrical energy. 
This is amplified and operates a galvanometer of special type, IV, 
so designed that the variations in electrical intensity are transformed 
into variations in the intensity of a beam of light. These variations 




LOUD SPEAKER 
CONVERTING 
ELECTRICITY 
INTO SOUND 



-PHOTOELECTRIC 
** CELL 




MICROPHONE 
CONVERTING 
SOUND INTO 
ELECTRICITY 



REPRODUCER CONVERTING THE 
VARIATION* IN TRANSMISSION 
Or THE FILM INTO ELECTRICAL 
ENERGY 





MODULATOR 
CHANGING THE 
NTENSITY OF A 
LIGHT IN ACCORDANCE 
WITH THE ELECTRICAL 
INTENSITY 



FILM RECORDING THE 
LIGHT INTENSITY ON 
THE SOUND TRACK 

OF THE NEGATIVE. 




PRINTING THE NEGATIVE 
ONTO THE POSITIVE FILM 
CONTAINING THE PICTURE 



FIG. 2. Cycle of operations used in sound recording. 



are recorded on a moving film at V. The film is developed, VI, and 
printed on a positive film, VII, which is developed and then in the 
projector controls the intensity of a beam of light from a lamp, IX. 
The light transmitted by the film falls upon a photoelectric cell and is 
converted into electrical energy. This energy, after amplification, 
operates the loud speaker, XI, which reconverts the electricity into 
sound. The reproduction thus takes place through six transforma- 
tions: The sound is (1) converted into electricity, the modulations 



288 



C. E. K. MEES 



[J. S. M. P. E. 



of which are (2} transformed into variations of light intensity; this 
then (3) undergoes the chemical transformation of the photographic 
process, which takes place in the four stages V, VI, VII, and VIII; 
in Stage IX, (4) the silver deposit produces changes of light intensity, 
which are (5) transformed into electrical variations, and these, finally, 
(6) into sound by the loud speaker. 

For ideal reproduction, the sound reaching the audience, XII, 
should be of exactly the same character as that reaching the micro- 




FIG. 3. Cycle of operations in tone recording. 

phone, //, the intensity, of course, being adjusted by appropriate con- 
trol of the amplifier. The sound will consist of a complex mixture of 
simple sounds varying particularly in the frequency and relative in- 
tensity of the different components; and in order that the quality of 
reproduction shall be satisfactory, it is necessary that the relative 
intensities of the different frequencies, and also the actual form of the 
pressure waves shall be preserved within certain limits through the 
series of transformations. This involves many problems in the 
design of the acoustical and electrical apparatus and circuits, but 



April, 1935] SOUND RECORDING 289 

the discussion of these is beyond the scope of this paper. I need only 
state that within certain limits, which limits may be regarded as 
very satisfactory for most practical purposes, the necessary conditions 
can be fulfilled. This paper is concerned with the conditions 
which are necessary for the operation of stages V, VI, VII, and VIII; 
that is, with the properties of the negative and positive photographic 
films and with the conditions under which they must be used in 
order to ensure satisfactory reproduction of the sound. 

Now, the positive film which carries the sound-track from which 
reproduction is achieved also carries the pictures showing the move- 
ment with which the sound is synchronized; and in order that the 
pictures shall be of satisfactory quality, it is necessary that certain 
conditions shall be achieved. 1 The cycle of operations necessary to 
produce a photographic reproduction is depicted in Fig. 3, which I 
owe to Dr. L. A. Jones. In the top right-hand corner will be seen a 
black cross illuminated by the sun, which will serve as a symbol for 
any photographic subject. This is reproduced in the left-hand top 
corner as an image on film for projection. The accuracy with which 
the picture duplicates in the brightnesses of its tones the different 
parts of the original subject corresponds to what is generally called 
the objective phase of tone reproduction. 

Let us consider the steps of the photographic process by which the 
object is translated into the print. The first step is the projection of 
the object by means of a lens upon the sensitive material, where it 
forms an optical image and, after the duration of a given time, pro- 
duces a latent image. This is developed, the material is fixed, 
washed, and dried, and we obtain a negative. This negative is 
printed either by contact or projection upon the positive material, 
producing first an optical image, then a latent image, and then, upon 
development, a positive print. 

When the image falling upon the sensitive material is transformed 
into a negative, the accuracy of the tone reproduction depends upon 
the shape of the well-known curve shown in Fig. 4, which represents 
the growth of density hi a photographic material as the exposure is 
increased. This shows what is known as the characteristic curve of an 
emulsion. There are three fairly well defined regions of the curve: 
Thus, from A to B we have the initial part, convex to the log E axis, 
which may be termed the region of underexposure*', between B and C, 
known as the region of correct exposure, the increase of density is 
practically constant for each increase of exposure; and in the third 



290 



C. E. K. MEES 



[J. S. M. P. E. 



region, from C to D, this arithmetical increase fails, until the density 
becomes constant; this is the region of overexposure. By prolonga- 
tion of the straight-line portion of the curve, the log E axis is cut at a 
point which Hurter and Driffield termed the inertia, which, when 
divided into a factor, gives the speed of the film. The slope of the 
straight line is known as 7 (gamma). Tone reproduction will be 
correct only over the straight-line portion of the curve, and in order 
to get correct tone reproduction, therefore, the straight-line portion 
must be sufficiently long to cover the entire scale of brightnesses oc- 

























TLA 

2.1 
1.8 
1.5 
12 
.9 
.6 
3 

0. 








































D 
















* 


^ 
















. 


? 


















/ 


















/ 


















/ 


















y 














A 


^ 


y 
















i 4 : 


l.O .3 .6 .9 2.1 .5 .8 3. 
LOG E. 



INERTIA 

FIG. 4. Characteristic curve of negative material. 

curring in the subject, and the exposure must be such as to place the 
lowest tone of the object at the beginning of the straight line of the 
characteristic curve. With modern materials and careful judgment 
of exposure, there is no difficulty in accomplishing this; and it may 
be said that there is no difficulty, provided the exposure is calculated 
correctly, in obtaining a negative in which the scale of brightnesses of 
the original will be rendered in correct proportion as a corresponding 
density scale in the negative. The absolute value of this scale de- 
pends upon the time of development of the negative, the contrast 
increasing as the time of development is prolonged, so that the scale 



April, 1935] 



SOUND RECORDING 



291 



may be contracted or expanded in the negative by varying the time 
of development without, however, affecting the proportional repro- 
duction of the relative brightness values. 

When negatives are developed for various lengths of time, the 
straight-line portions of the characteristic curves usually intersect 
approximately at the inertia point (see Fig. 5), so that the differ- 
ences between them can be expressed simply in terms of the numeri- 
cal value of 7. If, now, the value of 7 be plotted against the time of 
development, an exponential curve is obtained, of which an example 
is shown in Fig. 6. The maximum value of 7 is known as 7^ 
(gamma infinity), which is the 7 obtained when development is in- 



4 

2.1 


















. 


^~~' 


















f 




















/ 






/ 




1.5 

09 
06 
03 


( 












/ 


' 




S 














J 


/ 


/ 


/ 
















/ 


/ 


' 
















/ 


/\ 




















X 




















^ 


















U 04 1 0.7 1.0 \3 \.6 1.9 tZ 2.5 2.6 3.1 3. 
LOG I LOO t 



FIG. 5. Characteristic curves of negatives developed for 
two different times. 



definitely prolonged. In practice, it is convenient to express the 
contrast of a given material in terms of 7 > and of the 7 obtained by 
a fixed time of development in a given developing solution. 

In making a motion picture, the negative is printed upon positive 
film, for which the characteristic curve is shown in Fig. 7. The 
straight-line portion of this curve is of sufficient length to cover the 
full range of brightnesses occurring in practically all scenes photo- 
graphed, so that, provided exposure and development are correct, 
the brightness of the tones in the projected picture will be propor- 
tional to that of the tones in the scenes reproduced. We can follow 
through the tone reproduction by means of a diagram designed by 



292 



C. E. K. MEES 



[J. S. M. P. E. 



Dr. Jones (Fig. 8). In the right-hand lower quadrant of this dia- 
gram is plotted the characteristic curve of the negative material. 
Along the top of this quadrant are plotted the brightnesses occurring 
in the subject. In order to make the operation plain, two values of 
brightness have been selected and are indicated by broken lines; 
these produce corresponding densities upon the negative material. 
In the left-hand bottom diagram, the characteristic curve of the 
positive film is plotted, with its density at the bottom of the dia- 
gram and its exposure scale on the left-hand side. The original 
brightnesses translated into densities in the negative can now be 
transferred on to this curve by the broken lines shown. They thus 
appear as brightnesses of the print. In order to compare the bright- 



. O r 



1.5 



1.0 



-5 




o zo 40 GO so 

TIME IN MINS. CURVE CAl-C. 



100 



IZO 



FIG. 6. Relation between gamma and time of development. 

ness values of the print with those of the original scale, we draw new 
lines at right angles and transfer the reproduction of the brightnesses 
obtained through the two characteristic curves into the right-hand 
top quadrant, using the line C, drawn at 45 degrees in the left-hand 
top quadrant, to transfer the values to the point where they reach the 
vertical lines drawn from the original brightnesses. The intersec- 
tions of all the original brightness values with the reproduced bright- 
nesses of the print transferred in this way plot out the curve D, 
which represents the accuracy of reproduction of brightnesses 
throughout the photographic process. The straight-line portion of 
this corresponds to correct reproduction, and the curved portions to 
the errors introduced in the reproduction by the curved portions of 
the characteristic curves of the materials used. 



April, 1935] 



SOUND RECORDING 



293 



In practical sensitometry, as the study of the characteristic curves 
of photographic materials is called, it is usual to measure the density 
by diffused illumination. The density was defined by Hurter and 
Driffieldas: 



D = lo - 



A 

where T is the transmission; /, the light incident; and /i, the light 
transmitted by the silver deposit. Now, a silver deposit not only 




l.O 



FIG. 7. Characteristic curves of positive film. 

absorbs light it also scatters it; so that if a parallel beam of light 
falls upon such a deposit and passes on to a lens, part of the light is 
lost to the lens by absorption and part by scattering (Fig. 9). The 
density of the deposit 

D\\ = log fiU 

where D\\ is the density measured by parallel light of intensity 7||, 
and the transmitted light reaching the lens has the intensity I\. If, 



294 



C. E. K. MEES 



[J. S. M. P. E. 



on the other hand, the density of the deposit is measured by com- 
pletely diffuse illumination (as by placing opal glass in contact with 
the density) and the total light transmitted is effective (as in contact 
printing), then 



where D# is the density measured by diffuse illumination, I# is the 
incident diffuse illumination, and I\\ is the total intensity of the light 
transmitted. 

D\\ 
Callier 2 introduced this terminology, and the ratio -~ is known 




FIG. 8. Graphic solution of tone reproduction. 

as the Callier coefficient, Q. For motion picture positive film, Q has a 
value of approximately 1.4. 

In order that the reproduction shall be exact for the straight-line 
portion of the curve, and not merely proportional, the condition to 
be fulfilled is 

Tneg. X Tpos. = 1 

provided that the densities in all cases are measured in the same way. 
In motion picture photography, however, 7 neg> is measured by diffuse 
illumination since the negative is used for contact printing, but the 
7 of the positive effective in projection is increased because the posi- 



April, 1935] 



SOUND RECORDING 



295 



live is illuminated by a condenser and projected by a lens, so that 
this equation should be corrected to read 

Tneg. X Tpos. = j-^ 

where 1.4 is the value of Callier's Q for positive film and the density 
measurements are made by diffuse illumination throughout. 

In motion picture practice, it is found that this produces too flat 
a picture, owing to flare in projection and some psychological factors, 
and it is usual to give the virtual or over-all gamma (7 ne g. X 7 pos . ) a 
value of 1.2 (densities being measured throughout by diffuse illumina- 
tion). In order to keep graininess at a minimum, it is usual to de- 
velop the negative to a gamma of 0.60 and the positive to a gamma 
of 2.0. 




FIG. 9. Illumination of a positive for projection. 

In the reproduction of the sound according to the cycle shown in 
Fig. 2, stages V, VI, VII, and VIII are the photographic operations 
with which we are concerned. These involve the following factors 
which are at our disposal for the control of the process : 

(A) The choice of the film used for making the sound record. 

(B) The amount of exposure given to the film. 

(C) The development of the sound record. 
CD) The choice of the positive film. 

() The exposure used for printing the sound record upon the positive film. 
(F) The development of the positive film. 

Of these factors, D and F are involved in the processes used for the 
production of the picture, and must therefore fulfill the conditions 
necessary for the correct reproduction of the tones of the picture. 
They are consequently more or less fixed, and are not free variables 
available for the control of the sound record. For this purpose, 
there are available, therefore, factors A, B, C, and E, the positive 



296 



C. E. K. MEES 



[J. S. M. P. E. 



film being of the standard type of emulsion known as motion picture 
positive film and its development being normally such as to give a 
gamma of 2.0, these conditions being set by the practice of the mo- 
tion picture industry for the production of pictures. 

Before considering the factors mentioned and their relation to the 
practice of sound recording, certain general matters must be dealt 
with. In the first place, there are in use two distinct types of sound 
record: These are known as the variable-density and variable-width 
types, respectively. 

The variable-density type of record is illustrated in 
Fig. 1. In this type, the density is uniform across 
the sound-track but is made to vary along the length 
of the track by variations in exposure which corre- 
spond with variations in the pressure produced by the 
sound at the microphone. A glow-lamp or a string 
galvanometer operating as a light modulator is a famil- 
iar example of suitable apparatus for this purpose. 

In the variable-width type of record, the exposing 
light is of constant intensity and the sound record, 
therefore, of constant density; but the width of the 
track illuminated varies in accordance with the sound 
pressure at the microphone. An oscillograph whose 
mirror is adjusted to illuminate half of the sound-track 
when there are no sounds at the microphone will pro- 
duce this type of record (Fig. 10). 

The physical phenomenon which we term sound, 
consists in rapid variations in atmospheric pressure. 
These pressure variations are ordinarily produced 
by the vibration of mechanical parts or by the vibra- 
tion of columns of air, as in wind instruments. If 
the rate of vibration lies between about 16 and 20,000 
cycles per second, the result is a note of audible frequency. * From 
a mathematical standpoint, the simplest type of variation in sound 
pressure is a simple sine wave, as is shown in Fig. 11. This may be 
represented by 




FIG. 10. 
Variable- 
width sound- 
track. 



p = P + P sin 



(1) 



* The exact limit of audibility is dependent upon the intensity of the source and 
the characteristic of the observer's ear. For most practical purposes, it is suf- 
ficient to reproduce all frequencies between 20 and 10,000 cycles per second. 



April, 1935] 



SOUND RECORDING 



297 



where PO represents the average atmospheric pressure upon which is 
superimposed a sinusoidal variation of amplitude P. Although a 
simple sine wave of this sort is rarely produced by any musical in- 
strument, we may always analyze any musical note into a funda- 
mental sine variation plus a number of harmonics or overtones.* 
Each of these, however, is itself a simple sine wave whose rate of 
variation is an integral multiple of the fundamental. The quality, 
or timbre, of a musical instrument depends upon the relative ampli- 
tudes of these harmonics. 

If a perfect microphone and amplifier are used in recording, the 
current i in the last stage of the amplifier may be represented as 

i = /o + / sin at (2) 



1.0 


_ 


^ 








SIN Wl 

SCAL-O 

c 




/ 


\ 






/ 


<L > D 


/ 


^ 






/ 


fe? 






\ 




? 


ss 

^5 % 






\ 


/ 




-0* 






V 


^ 




-1 O 













O ^ PERIOD ^PERIOD /^PERIOD I PERIOD 

VALUES OF TIME, 

FIG. 11. A simple sine wave. 

There is a concept in sound recording which is very useful and 
which arises from this equation. Since the maximum value of sin otf 
is +1 and its minimum value 1, the maximum value of * is /o + 1, 
the minimum value is / 1 , and the average value is J . The modu- 
lation, m, is defined as 

* Consider the case of a musical instrument playing the A below middle C. 
The fundamental frequency of this note is a little more than 200 cycles per second, 
for simplicity, let us assume it to be exactly 200 cycles per second. In general, 
this note will consist of the fundamental 200-cycle variation, plus a harmonic 
with a frequency of 400 cycles per second, plus a third harmonic with a frequency 
three times the fundamental, or 600 cycles per second, and so on, the amplitude 
falling off rather rapidly in the higher harmonics. 



298 C. E. K. MEES [J. S. M. P. E. 

m = *ma%. *av. = *av._^inm. = ^ 

J. av. *av. * 

Since the amplifier current can never be less than zero, the value of 
m can never exceed unity. 

If, now, the variations in the exposure of the film are proportional 
to the variations in the current I, the exposure of the film may be 
represented by equation 4 : 

e = EQ + E sin wt (4) 

where e is the value of exposure corresponding to a sound pressure, 
Po 5 EQ is the value of exposure when there is no sound before the mi- 
crophone ; and E is the maximum value of the sinusoidal variation of 
exposure. 

The condition for correct tone reproduction in the photographic 
phase of the problem is simply that the current in the photoelectric 
cell of the reproducer be expressible by an equation in the form of 
equation 2: 

i' = V + KI sin a>t (5) 

The constant term / ' in eq. 5 is independent of the constant term 
/o in eq. 2. However, the amplitude of the variable term in eq. 5 
must be proportional to the variable term in eq. 2, as indicated by the 
proportionality constant K; that is, the transmission of the positive 
sound-track must be linearly related to the exposure of the negative. 
The transmission of the positive sound-track will vary, therefore, from 
an average value, T Q , to a maximum value, T Q + T, and a minimum 
value of TQ T, and T/TQ is the modulation. 

In the case of a variable-width record, the modulation is propor- 
tional to the width of the track covered by the silver deposit, the 
average transmission being that where half the width is occupied; 
the maximum transmission where all, or nearly all, the track is clear; 
and the minimum transmission where all, or nearly all, is covered. 
This average transmission in either type of sound record exists when 
no sound strikes the microphone. The transmission of the film, 
however, is not absolutely constant owing to surface imperfections, 
such as dirt, dust, scratches, etc., and also to the granular structure 
of the silver deposit. These changes in transmission cause a certain 
minimum of modulation which, when converted into sound, is known 
as ground-noise. 



April, 1935] SOUND RECORDING 299 

Sound has three attributes, namely: 

(1) Loudness. 

(2) Frequency or pitch. 

(5) Wave-form, quality, or timbre. 

Now, for perfect reproduction of sound, the reproduction should 
be perfect in respect of all these three attributes. As to loudness, the 
maximum intensity can be achieved by simple amplification (within 
the capacities of the electrical system and the loud speaker), but this 
amplification will also increase the minimum intensity due to ground- 
noise; so that the intensity range is limited primarily by the ground- 
noise, and secondarily by the modulation which is permissible before 
the wave-form becomes sufficiently distorted to be noticeable to the 
ear. 

As to frequency, there is no photographic limit at low frequencies, 
but the reproduction of high frequencies is limited by the resolving 
power of the photographic film as well as by the optical and mechani- 
cal systems. As to wave-form, the reproduction will be dependent 
upon the fulfillment of certain specific requirements in the photo- 
graphic operations. 

In systems in which sound is reproduced by means of electrical 
circuits, it is usual to express the relative loudness or intensity in 
terms of the electrical transmission unit, the decibel. If, in a circuit, 
the electrical energy is diminished to one-tenth of its input level, the 
loss is stated to be ten decibels, the decibel being defined as one-tenth 
of the logarithm of p^o^r^uTput' Correspondingly, if one sound has 
twice the energy of another, it is said to be three decibels (10 X 0.30; 
i. e., 10 X log 2) louder; if it has ten times the energy, it will be ten deci- 
bels louder; one hundred times the energy, twenty decibels, and so on. 

A change of one decibel is just about the smallest change in loud- 
ness which the ear can recognize, i. e., about 12 per cent. The in- 
tensity of sound is usually stated as the number of decibels above the 
audibility threshold, corresponding to an acoustical energy of about 
4 X 10 ~ 16 watts per square centimeter. A soft whisper at a distance 
of three feet would be 15 to 20 decibels above this threshold; speech, 
60 to 80 decibels ; the range of an orchestra from 40, for a single in- 
strument pianissimo, to 115 for the whole orchestra fortissimo* 

Now, as has already been mentioned, the maximum intensity of 
the sound produced from a given track can be increased to any extent 
within reason by increasing the amplification between the photo- 
electric cell and the loud speaker, but this amplification will proper- 



300 



C. E. K. MEES 



[J. S. M. P. E. 



tionately increase the minimum intensity of sound, or ground-noise, 
and the ratio between the maximum and the minimum intensity will 
remain constant. This ratio can, however, be increased by the meth- 
ods known as noiseless recording, which will be discussed shortly. 

The ground-noise is the sound heard from the loud speaker when 
the film is running through the projector, but there is no change in 
transmission of the sound-track except that due to parasitic modula- 
tion. When a well adjusted projector is running without any film at 
all and the amplifier is adjusted to give the desired loudness from a 



-25 



-30 



o 

z 

J-40 

y 

I 

U-45 



-50 



-56 



O riNE. GRMN EIMUL-<=>\ON 

- MEDIUM SVZ.E. GRfMN ELMUL.. 

X COURSE GR/MN EMULSION 




O.b 



1.0 



Z.S 



SPECULAR DENSITY 

FIG. 12. The relation between ground-noise and density. 

normal sound record, there is a slight residual noise arising from the 
photo-cell and electrical system, the level of which, however, should 
not be greater than 55 decibels below the maximum level of the sound 
record. If a perfectly clean film is allowed to run through the pro- 
jector, the ground-noise increases about 6 decibels. If, now, the film 
carries a photographic density such that it transmits one-quarter of 
the light (this is the density normally used for an unmodulated vari- 
able-density sound-track) and if the amplification is increased to 
compensate for the reduced illumination of the cell, the noise level 



April, 1935] SOUND RECORDING 301 

will increase three more decibels even if the utmost care has been 
taken to preserve the film-track from extraneous dirt. This increase 
in the noise level is due to the grain structure of the photographic 
deposit. 

Fig. 12 shows the relation between ground-noise and the specular 
density of the photographic deposit. It shows also that the noise 
increases with the increase hi the emulsion grain size. The ground- 
noise, however, increases when the film is run through the projector, 
as it accumulates microscopic scratches and specks of dust, and al- 
though cleaning will reduce the ground-noise somewhat, a film in 
active service in the theater which has been run some fifty times (on 
the whole, a favorable condition for a first-run film) has its average 
value of noise level raised by about 6 decibels. The apparent in- 
crease in ground-noise is much greater than is indicated by this 
figure because of its character. 

In view of this great interest in the ground-noise, due to unavoid- 
able handling of the film, any attempt to reduce the ground-noise due 
to the granularity of the silver deposit does not seem to be worth 
while, in view of existing theater conditions ; although if some way of 
avoiding the ground-noises due to other causes could be found, a re- 
duction of the granular structure of the film is not impossible and 
might in that case be of value. 

It is obvious that ground-noise is most noticeable when the signal 
level is low and is of negligible importance when the recorded sound 
is loud. Since the ground-noise is due chiefly to dirt, dust, scratches, 
oil spots, and finger prints on the film surface, and since these surface 
imperfections produce a larger percentage change in the photoelectric 
cell illumination through a film which has a high transmission than 
through one which has a low transmission, it is evident that the 
ground-noise could be diminished if the average transmission of the 
film decreased with a decrease in the sound level. In the case of 
variable-density recording, this could be accomplished if the positive 
sound-track were denser when the modulation was low and became 
lighter as the sound increased. This method is utilized in the West- 
ern Electric system of "noiseless" recording. 

In this system of recording, the exposure on the negative film is 
made through a light-valve whose ribbons are normally spaced 0.001 
inch apart, giving a certain fixed unmodulated density in the sound- 
track of the negative and, in turn, of its print. This ribbon spacing 
is sufficient to permit the movement of the ribbons required by the 



302 



C. E. K. MEES 



[J. S. M. P. E. 



loudest sounds and is, therefore, considerably greater than necessary 
for the weaker sounds. 

In noiseless recording, the mean spacing of the ribbons is not con- 
stant but is reduced to some predetermined value during silent pas- 
sages or during periods of weak signals. This reduces the density of 
the negative unmodulated track and, consequently, increases the 
density of the positive unmodulated track, decreasing the ground- 
noise. As louder and louder sounds arrive at the microphone, in- 
creasing the valve modulation, the mean spacing of the ribbon in- 
creases sufficiently to accommodate the increased modulation of the 
valve without bringing the ribbons together. 

Fig. 13(5) illustrates the behavior of the ribbons in a normal 
light-valve. In the normal method of recording, the ribbons have 



C NOISELESS RCDG 



B NORMAL RCDG. 



A SPEECH INPUT 



FIG. 13. Diagram illustrating the use of "noiseless recording." 

a constant average spacing and their movement is essentially simple, 
corresponding to the variations of the sound-current only. In the 
method of noiseless recording, the ribbons may be regarded as having 
two motions: first, the motion due to the sound-currents only, ex- 
actly as in the normal method of recording; and, second, a superim- 
posed, slower movement which follows the envelope of these sound- 
currents. This is shown in Fig. 13(C). This control of the ribbons 
is effected by a special electrical circuit which supplies to the ribbons, 
not only the sound-currents, but another current at a slower period 
corresponding to an average value or envelope of the sound-currents, 
which are shown in Fig. 13(^4). As a result, the exposure through 
the light-valve on the negative film is reduced during periods of si- 
lence, while at the same time provision is made for increasing the ex- 




April, 1935] 



SOUND RECORDING 



303 



posure automatically with increasing modulation of the light-beam 
by the light- valve ribbons. It follows that the density of the result- 
ing negative sound-track will be a minimum during silent intervals 
and will rise to a maximum value with increasing input, while the 
density of the print made from this negative will be a maximum 
during silent intervals and will decrease to a fixed minimum with 
increasing output from the film. By this system, an effective noise 
reduction of approximately 10 decibels can be attained. 

Turning to the variable-width system 
of recording, the same result is achieved 
if the width of the clear or unexposed 
portion of the positive sound-track is 
less when the modulation is low and be- 
comes wider as the modulation increases. 
In the latest type of RCA recording 
system, the galvanometer is so arranged 
that when a current such as the ampli- 
fied microphone current flows through 
it, its mirror vibrates about a horizontal 
axis. Associated with this galvanometer 
mirror are a source of illumination and 
an optical system which image an il- 
luminated equilateral triangular area in 
the plane of the slit. The base-line of 
this triangle is parallel to the major axis 
of the slit, and a perpendicular line 
drawn from the apex of the triangle to 
its base passes through the midpoint of 
the slit. The height of the illuminated 
triangle is large compared to the height 
or the width of the slit opening, so that 
at any one time only a small fraction of the light falling within the 
boundary of the illuminated triangle passes through the slit opening. 

For normal recording, the mean position of the triangle is adjusted 
so that the center of the slit opening lies at a distance from the apex 
equal to one-half its maximum amplitude. Thus, when no sound 
strikes the microphone, the negative receives a uniform exposure. 
The strip of the negative exposed is centrally located in the sound- 
track, and its width is equal to half that of the fully modulated sound- 
track. When sounds strike the microphone, the triangular area vi- 




(A) (B) 

FIG. 14. Normal and 

"noiseless" variable -width 
tracks. 



304 



C. E. K. MEES 



[J. S. M. P. E. 



brates up and down; thus illuminating more or less of the slit open- 
ing, causing a corresponding increase and decrease in the length of the 
slit image upon the film, with the corresponding widening and narrow- 
ing of the exposed area of the sound negative, which when developed 
and printed, results in the sound-track shown in Fig. 14(^1). In 
the case of normal recording, the average position of the triangle is 
fixed, and the exposure through the slit opening results in a clear 
portion in the positive sound-track sufficiently wide to accommodate 
full modulation, and it is therefore much wider than that required for 
low modulation. 

It is entirely permissible to reduce considerably the width of the 
clear portion of the track during periods of no sounds or weak sounds 

if, in the presence of louder sound, 
the width of the clear portion of 
the track is in some manner in- 
creased sufficiently that the am- 
plitude of the sound record does 
not exceed the width of the clear 
track. This is accomplished by 
adjusting the position of the illu- 
minated triangle with respect to 
the slit so that when the modula- 
tion is zero, the apex of the tri- 
angle extends just beyond the slit 
opening. Thus, the negative re- 
ceives a very narrow line of ex- 
FIG. 15. Curve showing the sharpness 

of a record at an edge. P osure alon S the middle of the 

track which, when developed to a 

negative and printed, results in the sound-track in Fig. 14(.B). Now, 
as the modulation increases, the position of the triangle changes so 
that the mean position of its apex is farther and farther away from 
the slit opening. When the modulation is complete, the distance be- 
tween the slit opening and the mean position of the apex is equal to 
one-half its maximum permissible amplitude. 

It will be seen, therefore, that in the normal method of recording 
the illuminated triangle has constant average position and its move- 
ment is essentially simple, corresponding to the variations of the voice- 
currents only. However, in the method of noiseless recording, the 
triangle may be regarded as having two motions: first, the motion 
due to the sound-current only, exactly as in the normal method of 



0.5 



DISTANCE IN MICRONS 



April, 1935] SOUND RECORDING 305 

recording; and, second, a superimposed, slower movement which 
follows the envelope of these sound-currents. 

THE REPRODUCTION OF FREQUENCY 

The extreme range of frequencies audible to the normal ear is from 
approximately 16 cycles per second to nearly 20,000 cycles per second. 
The last two octaves, however, are of very little importance when 
considered as pure sounds ; that is, a simple wave having a frequency 
greater than 5000 cycles is rarely met with, but frequencies between 
5000 and 10,000 cycles are of considerable importance as representing 
the harmonics (or overtones) of sounds of lower frequency and thus 
as modifying the character of the sound. In speech, for instance, 
the sibilants and especially the consonants can be reproduced correctly 
only if frequencies in this range are available. The older type of 
equipment used in the theaters, and especially the loud speakers, 
limited the maximum frequency in reproduction to 5000 to 6000 
cycles. At this frequency loss due to insufficient resolving power in 
the film is practically negligible. More recently, however, both the 
recording and the reproducing equipment have been improved so as 
to operate effectively with frequencies up to 10,000 cycles. This 
high-quality reproducing equipment has already been installed in 
many of the first-class theaters, so that the use of a photographic 
material of somewhat higher resolving power would be desirable. 

The resolving power of a photographic material is limited by the 
scattering of the silver halide grains for the incident light. Consider 
a beam of light incident upon a film and limited by a sharp edge. The 
light falling upon the film will be reflected from the silver halide 
grains and scattered into the shadow, the extent of the scattering de- 
pending upon the reflecting power of the grains, their size, and the 
absorption of the emulsion for the scattered light. The edge of an 
image, therefore, is not infinitely sharp, and the sharpness of the de- 
veloped image at an optically sharp edge takes the form shown in 
Fig. 15, which is of the same form as the characteristic curve of the 
emulsion. Sharpness is defined numerically as the slope of the 
straight-line portion of the sharpness curve. 

In practice, the resolving power depends not only upon the sharp- 
ness but also upon the graininess of the emulsion, and is usually mea- 
sured directly by photographing a series of very small grating test 
objects, the resolving power being expressed in terms of the number 
of lines per millimeter which are visibly resolved when the image is 



306 



C. E. K. MEES 



[J. S. M. P. E. 




FIG. 16. Photomicrographs showing the 
measurement of resolving power. 

examined under the microscope (Fig. 16). The resolving power of 
an emulsion is of an extremely complex nature, depending upon a 



April, 1935] 



SOUND RECORDING 



307 



number of variables such as distribution of intensity and contrast 
in the object, density of the photographic image, time of development, 
type of developer, quality (spectral composition) of the exposing radia- 
tion, and several other minor factors. It is impossible, therefore, 




i.o to so 40 

FIG. 17. Variation of resolving power with density. 

with our present knowledge at least, to express the resolving power 
of an emulsion in terms such that one can calculate very closely the 
depression in volume of a sound record incurred by imperfect resolu- 
tion. It may prove useful, however, to consider the nature and the 
effect of some of the above-mentioned variables. In order to sim- 




CONTRAST 



FIG. 18. Variation of resolving power with object contrast. 

plify the problem as much as possible, we shall consider each variable 
separately : 

Let us consider, first, the effect of a change in the image density 
keeping all other factors as nearly constant as possible. Fig. 17 



308 



C. E. K. MEES 



[J. S. M. P. E. 



shows a typical curve for motion picture positive film, the object 
being 1000; that is, the amount of light transmitted through the clear 
spaces is about 1000 times the amount of light transmitted through 
the opaque spaces, and the development time, 8 minutes in a standard 
positive film developer, this being a rather full development. This 
curve shows that the resolution increases from zero, at a density of 
zero, to a maximum value of 80, at an image density of 1.3. This 
density we shall call the optimal density. The general character of 
this curve is similar for any type of emulsion, although in general the 
lower the inherent contrast of the emulsion, the lower is the optimal 
density. Keeping the time of development constant and varying 
the object contrast, if we measure maximum resolving power, that is, 
resolving power at optimal image density, we obtain the curve shown 

in Fig. 18. The resolution in- 
creases with the contrast, finally 
reaching a limit. It is seen that 
motion picture positive emulsion 
very nearly reaches its maximum 
value when the object contrast 
is 1000. If, now, we vary the 
time of development, keep the 
object contrast 1000, and deter- 
mine the resolving power for each 
time of development at an image 
density of 0.3, we obtain the curve 
shown in Fig. 19. It is interest- 
ing in this case to note the rapid 

decrease in resolution with increasing time of development. This is 
because the optimal density is greater than 0.3 and the entire curve 
shifts to the right as the time of development increases, thus depres- 
sing the low-density end. If both the image density and the devel- 
opment time vary, we obtain a family of curves, and if the object 
contrast varies as well, the resolving power is represented by a family 
of complicated surfaces. 

To determine the resolution obtainable in a given practical case 
requires an integration of the effects due to these several variables. 
This is not possible, with our present knowledge, at least. It is prob- 
ably a fair assumption to make, however, that an emulsion having a 
maximum resolution of 80, when correctly used with a high-quality 
optical system, will give excellent rendering of tone and form at fre- 




FIG. 19. Variation of resolving power 
with time of development. 



April, 1935] 



SOUND RECORDING 



quencies well above the highest frequencies now used, and that the 
volume at those frequencies will not be materially reduced due to 
lack of definition. As a matter of fact, it is possible to record fre- 
quencies of 13,000 to 15,000 cycles and obtain excellent definition. In 
order to do so, the adjustment of focus has to be carried out with the 
utmost care. A displacement 
of the objective lens imaging 
the slit upon the film by a small 
fraction of a millimeter causes a 
change from excellent definition, 
that is, a high modulation in 
the density, in one case, to a 
practically uniform density in 
the other. Fig. 20 is a photo- 
micrograph of a variable-area 
sound record of 10,000 cycle fre- 
quency. The definition is not 
perfect, yet the volume depres- 
sion at this frequency from that 
cause certainly would not be 
very large. 

While, however, the resolving 
power of the emulsion is suffi- 
cient to record very high fre- 
quencies, the spread of the 
image does involve some loss 
in modulation. This is illus- 
trated in Fig. 21, which shows 
the loss of output level (in deci- 
bels) plotted against the fre- 
quency, the relative output 
levels being measured by a mi- 
crodensitomer. In practice, a 
greater loss results owing to the 

width of the slit of the reproducer. A loss is also introduced if any 
slippage between the negative and the positive films occurs in the 
printer, which will happen unless the perforation spacing of the 
positive and negative films at the time of printing is exactly that for 
which the printer was designed. For the printer most generally used, 
the diameter of the printing sprocket is such that the negative should 




FIG. 20. 



Photomicrograph of a 10,000- 
cycle record. 



310 



C. E. K. MEES 



[J. S. M. P. E. 




-10 



-12 



-14 - 



-16 
-18 h 



FREQUENCY N CYCLES PER SECOND 



FIG. 21. Loss of output level with increase in frequency. 

be 0.3 per cent below standard pitch when printed on standard posi- 
tive film. This represents the normal shrinkage of the sound nega- 



-z 



-4 



-6 



-8 



-10 



-IZ 



-14 



-16 



-18 




FREQUENCY IN CYCLES PER SECOND 



FIG. 22. Output plotted against frequency. 

tive film during the time which elapses between the processing of 
the negative and the printing of the bulk of the release positives. 
The over-all characteristic actually obtained in good practice, as 



April, 1935] 



SOUND RECORDING 



311 



measured by the output from a standard reproducer head plotted 
against frequency, is shown in Fig. 22. 

THE REPRODUCTION OF WAVE-FORMS 

In order that the original wave-form may be retained, it is neces- 
sary that the transmission of the positive sound-track shall be linearly 
related to the exposure of the negative. The photographic condi- 
tions can be analyzed by the tone reproduction diagram, for which 
graphic solutions have been discussed. Inasmuch, however, as the 
current of a photoelectric cell is linearly related to the transmission 
of the positive, it is convenient to use linear coordinates and to plot 



OVERALL. CHARACTERISTIC 

AVERAGE: TRANSMISSION 



NEGATIVE CHARACTERISTIC 

AVERAG.EL TRANSMISSION S0*7 
GAMMA O.6 




FIG. 23. Schematic reproduction diagram for sound recording. 

transmission against exposure instead of plotting density against the 
logarithm of exposure, as is customary in general photographic work. 
A schematic reproduction diagram is shown in Fig. 23. In quad- 
rant J, the transmission of the negative when developed to a gamma 
of 0.5 is plotted against the exposure. In quadrant //, the straight 
line is drawn at an angle which defines the exposure on the positive 
film in the printer. The characteristic curve of the positive film de- 
veloped to a gamma of 2.0 is plotted in quadrant ///, the transmission 
being plotted against the exposure. Vertical lines for each exposure 
point in the negative in quadrant / now meet horizontal lines drawn 
through the corresponding positive exposures on the positive curve 
in quadrant 177, and their intercepts trace out the curves shown in 
quadrant IV, which represents the reproduction curve of the process. 



312 



C. E. K. MEES 



[J. S. M. P. E. 



So far as this reproduction curve in quadrant IF is linear, undis- 
torted reproduction should be obtained. 

It is clear that the effect of modification in the photographic proc- 
ess, use of different film, different exposures, or different times of de- 
velopment can be computed, and this was done some years ago by 
several workers; but in view of the complexity of the problem, it 
seemed desirable to make a direct experimental study of the subject. 
To this end, three special instruments were designed. 




FIG. 24. Variable-slit sensitometer. 

The first was a variable-slit sensitometer. The instrument is 
shown in Fig. 24. It consists of a camera moving the film at a uni- 
form speed of 90 feet per minute past an exposing slit of variable 
width. The light-source is a tungsten ribbon filament lamp. The 
lamp filament is imaged upon the variable-width slit, which in turn 
is imaged upon the moving film at two to one reduction, as shown in 
Fig. 25. The width of the slit is varied automatically by the rota- 
tion of a precision screw which is actuated by a cam mechanism. 
There are sixteen stops on the cam, so arranged that the width of the 
variable slit is changed by steps from a slit-width of 4 mils to about 



April, 1935] 



SOUND RECORDING 



313 



0.016 mil. This range of variation is more than sufficient to cover 
the normal exposure range occurring in practice. With this instru- 
ment, film can be exposed under exactly the conditions of time of ex- 
posure and intensity of light which prevail when sound is recorded. 

The second instrument is a sound-track densitometer, shown in 
Fig. 26. This instrument is so designed as to approximate closely 
the conditions of a standard sound reproducer except that the film 
instead of being moved at a speed of 90 feet per minute by a rotating 
sprocket, is moved step by step past the scanning slit by the rotation 
of a very accurate screw mechanism. This instrument is used chiefly 
to measure densities photoelectrically; since the ratio of radiation 




FIG. 25. Diagram of optical system of slit sensitometer. 

transmitted to that incident upon a given photographic image de- 
pends upon the degree of collimation or diffusion of the incident radia- 
tion and upon the size of the cone angle of the transmitted radiation 
intercepted by the receiving element, it is necessary in connection 
with the sensitometry of sound recording emulsion that densities 
should be measured to conform with the optical conditions which 
obtain in practice, the densities attained being of the same order as 
those defined by Callier as D\\ t and not the D# values used in or- 
dinary sensitometry. 4 

The third instrument is a microdensitometer of rather special de- 
sign, by which continuous microdensographs can be obtained from 



314 



C. E. K. MEES 



[J. S. M. P. E. 



the sound-track. In this instrument, a microscope objective pro- 
jects an enlarged image of the sound-track upon a slit, behind which is 
a lens imaging the principal plane of the microscope upon the photo- 
electric cell surface. This insures a uniform distribution of illumina- 
tion on the photoelectric cell surface at all times. The cell operates 
into a strictly linear d-c. amplifier whose power output actuates a 
galvanometer of fairly low period. The microscope stage carrying 
the film is driven at a constant speed by means of a train of gears and 
a precision screw. The image of the sound-track moves across the 




FIG. 26. Diagram of sound-track densitometer. 

slit, varying the illumination of the photoelectric cell and, therefore, 
its current output, in accordance with the transmission of the sound- 
track. The amplified photoelectric current actuates the galvanome- 
ter whose angular movements are recorded upon a continuously 
moving strip of film 6 l / 2 inches wide and driven at a constant ratio 
of speed with respect to the sound record. The ratio of sound-track 
speed to recording film speed is adjusted so that the amplitude and 
the wavelength of the record are each about five inches, a convenient 
size to handle and sufficiently accurate for all purposes. One such 
record is shown in Fig. 27. Superimposed upon the curve are shown 



April, 1935 J SOUND RECORDING 315 

points obtained when the same record is run through the sound-film 
densitometer containing a standard reproducer optical system and 
photoelectric cell. The photoelectric cell current can then be read 
directly on the galvanometer at various points along the curve. The 
results of measurements made on the two instruments agree well. 

The results obtained with these instruments are studied by the use 
of a harmonic analyzer of the Henrici type, made by Coradi. It 
normally gives the first five terms of the Fourier series, although the 
next five terms may be obtained by a second set of readings, giving, 
in all, the first ten harmonics. A complete description of this in- 
strument is given by D. C. Miller. 6 

The microdensitometer records are enlarged to a standard wave- 
length of 40 centimeters, this being the required base-line for the 




FIG. 27. Microdensitometer record. 

harmonic analyzer. The results are accurate with a single tracing 
of the curve to the order of 0.5 mm. of amplitude for the lower har- 
monics, and a slightly greater accuracy can be attained for the higher 
terms because the repetition of the values gives a better average. 
Since the curves are about 300 mm. in amplitude, the harmonics are 
determined with a maximum error of about 0.2 per cent. In order 
to decrease accidental errors, the average of four successive cycles 
was used in most of the work. This was found to be sufficient to 
give the required accuracy of analysis of the wave-form even when 
the modulation was low. In order to correlate the two fundamental 
quantities of modulation and harmonic content, the amplitude was 
measured on the identical cycles of the wave. 
Fig. 28 shows the results of analyzing a very distorted sound-wave, 



316 



C. E. K. MEES 



[J. S. M. P. E. 



the negative having an average density of 1 .00 (diffuse) at a gamma 
of 1.25. The print shown has an average density of 0.60 and on 
analysis was found to have the relative amplitudes of harmonic shown. 
Components as far as the sixth harmonic have been drawn, although 




FIG. 28. 



Analysis and synthesis of sound 
record. 



the amplitudes of the fifth and sixth are only 0.7 per cent that of the 
fundamental, corresponding to 43 decibels down. 

The tone reproduction diagram shown in Fig. 23 and referred to 
briefly above will now be considered in more detail. In quadrant /, 
the exposure follows along the axis from an average value indicated 



April, 1935] 



SOUND RECORDING 



317 



by the broken line. The exposure varies equally in the direction of 
increasing and decreasing exposures. The average or unmodulated 
exposure results in an average transmission, while the maximum and 
minimum values depend upon the modulation. Using the arbitrary 
value of 20 as average exposure, the maximum modulation occurs 
when the light decreases to zero on the decreasing half of the cycle 
and increases to 40, on the increasing half. In terms of the light- 
valve, this maximum means that the valve ribbons close completely, 
or clash, on the one hand, and open to double their average value, 
on the other. In terms of a glow-lamp, the lamp is extinguished and 



POSITIVE: 
CHARACTERISTIC 

GAMMA =20 



QUADRANT XH 



O 2O /4O GO BO 

posmve /EXPOSURE: 



PRINTER 
CHARACTERISTIC 



QUADRANT DL 



Z 
o 

Seo- 

I- 

IT 



80- 



60 



40 



20 



OVERALL CHARACTERISTIC 



GAMMA PRODUCTS 



f A = 
^ B= I 

u- 



O.7 
O 



UNMODULATED TRANSMISSION 
30-7. 



QUADRANT 1ST 





NEGATIVE CHARACTERISTIC 

GAMMA 
A - 0.35 
B - 0.^0 
C*0-80 
QUADRANT I 



4O 6O 8O IOO fZO 

NEGATIVE EXPOSURE: 



FIG. 29. Effect of negative development upon reproduction. 



goes to double its average intensity on the two halves of the cycle. 
In general, the amplification between the sound-waves striking the 
microphone and the recording device is so adjusted that this maxi- 
mum is never reached, because the limits of the curve can not be 
utilized without introducing a considerable amount of distortion. 
If the modulation is limited to 80 per cent, corresponding to about 2 
decibels below the overload point, the remaining distortions can be 
balanced out, as can be seen by following through the diagram. 

The straight line in quadrant II defines the positive exposure. 
The exposure of the positive material used for printing will be pro- 



318 



C. E. K. MEES 



[J. S. M. P. E. 



portional to the transmission of the negative, the coefficient of pro- 
portionality depending upon the intensity of the exposing lamp. 
Zero transmission of the negative always leads to zero exposure of the 
print, so a family of straight lines through the origin represents all 
possible printing conditions. 

Quadrant 177 shows the positive characteristic drawn for a gamma 
of 2.0. This is similar to the negative curve. 

Quadrant IV shows the final curve of positive transmission against 
negative exposure. This curve is obtained by following negative ex- 
posures through to the corresponding positive transmission. As 



POSITIVE 
CHARACTERISTIC 

GAMMA "Z.O 



60 



40 



4O CO 8O IOO 
POSITIVE /EXPOSURE 



80 



CO 



PRINTER 40 

CHARACTER 1ST 1C 



ZO 



-^ OVERALL. 
CHARACTERISTIC 




NEGATIVE CHARACTERISTIC 
AVE. 

GAMMA DENSITY 
A - .55 .20 

- .55 .I* 




20 



4O GO 8O IOO IZO 
NEGATIVE, EXPOSURE. 



FIG. 30. Effect of negative density upon reproduction. 

can be seen, the final curve is relatively straight, the curvatures of 
the two photographic materials used having compensated for each 
other. 6 Now, since the photoelectric cell current in the reproducer 
is proportional to the transmission of the film, it follows that, for the 
photographic process to be correct, there must be a straight-line re- 
lationship between positive transmission and negative exposure. It 
is also true that any straight line gives correct sound tone reproduc- 
tion. The steepness of the line determines only the amount of modu- 
lation necessary in recording, while its vertical projection determines 
the amount of change in photoelectric cell current and hence the 
amount of voltage developed in the first stage of the amplifying sys- 



April, 1935] 



SOUND RECORDING 



319 



tern. Control of these factors does not lead to any distortion corre- 
sponding to that produced by a wrong over-all gamma in the pictorial 
case, where the picture becomes either too contrasty or too flat. The 
analog of pictorial contrast is loudness, which can be controlled by 
adjustment of the amplifiers. 

As can be seen from the figure and by comparison with the H&D 
curve, much of the negative and positive exposure range is in a region 
corresponding to the toe of the H&D curve, while a part is on the 
straight line. Since the relationship actually utilized is the linear 
one, it is difficult to interpret the results on the basis of considerations 



POSITIVE CHARACTERISTIC 
I GAMMA -2.O 



POS. EXPOSURE 




BO 



60 



40 



PRINTER 
CHARACTERISTIC 



OVERAUU CHARACTERISTIC 




AVE. DENS. 



8 .40 
C. 65 



NEGATIVE CHARACTERISTIC 

GAMMA 0.50 
AVERAGE DENSITY- O.C5 




4O CO 8O 100 IZO I4O 
NEGATIVE EXPOSURE. 



FIG. 31. Effect of positive transmission upon reproduction. 

of the straight-line portion of the curve. In actual practice, only 
peaks of waves reach to a high enough density to be on the straight 
line of the H&D curve of the positive. 

So far, the limitations on the straight line of quadrant IV have not 
been mentioned. There would be no problem of tone reproduction 
if it were not of importance to keep the amplification to a minimum 
in order to keep down the ground-noise. The amplification neces- 
sary depends directly upon the range of transmission used. This 
points to the desirability of using 50 per cent as the average trans- 
mission of prints, as the light modulation in reproducing can then be 
from per cent to 100 per cent; that is, the absolute maximum. 



320 



C. E. K. MEES 



[J. S. M. P. E. 



Consideration of ground-noise leads to the practical adoption of 
lower values of transmission, and these are compatible with satis- 
factory tone reproduction, as shown in the next figures. 

Fig. 29 shows the effect of variable negative development upon the 
final reproduction. Gamma is given on the curves to indicate ex- 
tent of development. It can be seen from quadrant IV that curve C 
has the greatest vertical height of straight-line portion, and this 
corresponds to the greatest negative development. In general, the 
change of quality with negative development is not very rapid. 



POSITIVE; CHARACTERISTIC CURVE. 

O-4% 7= tO 



POSITIVE EXPOSURE^ 



IS" 

VE WAVE: roR-\ 



TO. 
CHARACTERISTIC 



I 
! 



I- SINUSOIDAL. EXPOSUREL- 
ZEL-F~IISAL. WAVE 



-1C.HT VALVE MODULATION 



m 

PRINTING 

CK AR ACTER I %T 1C 





t. A & 6 IO IX 14 (ft IS 

FIG. 32. Reproduction of a sinusoidal wave. 

Fig. 30 shows the effect of varying negative density, the extent of 
development being constant. The printer characteristic was changed 
to adjust the different transmissions of the negatives to the same 
positive transmission. The negatives used are extremes, and an 
intermediate value would give fairly good sound reproduction, as the 
curve would lie between A and B in quadrant IV. Printing the low- 
density negative to a high positive density gives AB with practically 
no change in positive transmission and hence very low photo-cell 
current modulation, while printing the high-density negative to a 
low-density print gives BA , a record with sufficient change in trans- 



April. 1935] 



SOUND RECORDING 



321 



mission, hence sufficient volume, but very bad distortion, as indicated 
by the curvatures. 

Fig. 31 shows the effect of variation of positive transmission, the 
negative being constant, and the change made by varying the printer 
lamp intensity. The family of curves in quadrant IV shows a pro- 
gressive change with increasing positive density from a characteristic 
very concave to the exposure axis, A to D, which is convex to the ex- 
posure axis. Curve C is practically a straight line and would give 
satisfactory reproduction with a positive density of 0.65. The am- 



I- SINUSOIDAL EXPOSURE. 

vn-c-iNAt_ WAX/E 



J> NEGATIVE CHARACTERISTIC CUR V 




FIG. 33. Reproduction of a sinusoidal curve introducing distortion. 

plitude of reproduction measured by the vertical projection corre- 
sponds to 33 per cent modulation of transmission, which is also satis- 
factory from the standpoint of freedom from ground-noise. 

Fig. 32 shows graphically the steps in the photographic sound re- 
cording process as applied to a sinusoidal sound-wave. The cycle of 
operation begins in the right-central compartment with the solid 
line showing a simple sine wave of sufficient amplitude to modulate 
the light-valve 90 per cent; that is, from the normal value of 0.001 
inch to a value of 0.0001 inch on the low-exposure side, and to a value 
of 0.0019 inch on the high-exposure side. This modulation, in terms 



322 



C. E. K. MEES 



[J. S. M. P. E. 



of exposure units, results in the negative curve in the lower right- 
hand compartment. The actual wave-form of the negative is in 
compartment IV. The curved characteristic indicates the presence 
of large distortion. This distorted wave is printed on to the positive 
characteristic of compartment V. A straight line is used in compart- 
ment VI to adjust the amplitude of the print to equal that of the 
original wave. The flattening at the crest and the trough indicates 
the presence of some third-harmonic distortion. 

Fig. 33 shows the same type of diagram as Fig. 32. This shows 
that a combination of high negative density and long time of develop- 
ment leads to a considerably more distorted final wave-form. The 



r 

fin 

u 
a 



S2 
I 8 



^ 



<A 







.8 t.O IZ 1.4 1.6 IB 10 

PRINT DENSTV 



FIG. 34. Distortion as a function of print density in 
variable-width recording. 

excessive flattening in the region corresponding to low exposures in- 
dicates that in this record there is appreciable second-harmonic dis- 
tortion as well as third. 

By means of diagrams of this kind, the distortion in wave-form of 
the recorded sound can be determined. Actual analysis of many 
records shows that considerable latitude in the exposure processing 
conditions is allowable, as the harmonic distortion does not change 
rapidly with the change of these negative or positive conditions. The 
ear is not critical of small amounts of harmonic distortion. 

Turning to the variable-width method of recording, it can be seen 
that, with this type of record, faithfulness of reproduction is pri- 
marily a matter of reproducing a geometric form. The optical sys- 



April, 1935] 



SOUND RECORDING 



323 



tern is so adjusted that when no sound reaches the microphone, a 
part of the sound-track is illuminated to a fixed intensity, leaving the 
remainder of the track unexposed. When sound strikes the micro- 
phone, the moving part in the recorder optical system that is, the 
galvanometer mirror is caused to vibrate and vary the width of the 
illuminated portion of the track. The exposure of whatever fraction 
of the width of the track is illuminated is at all times constant. The 
wave-form of the recorded sound suffers a certain amount of distor- 
tion due to lack of sharpness at the boundary of the exposed area. 
This lack of sharpness is due in part to scattered light in the optical 
system, a fraction of which reaches the part of the sound-track lying 




FIG. 35. Modulation of cell current as a function of 
density in variable-width recording. 

without the geometric boundary of the image, and is also due to light 
scattered by the emulsion itself, giving rise to a finite gradient at the 
edge of the photographic image. 

Many analyses have been made of these records and typical re- 
sults are shown in Figs. 34 and 35. Fig. 34 shows the harmonic 
distortion as a function of print density for a series of negative den- 
sities. Electrical currents for modulating the galvanometer were 
derived from a beat frequency whose output was a pure sine wave. 
The curves show that over a wide range of negative and print ex- 
posures the total harmonic content is below the maximum allowable 
value. The curves show also that for each negative density there is 
a print density giving a minimum harmonic content. This mini- 



324 C. E. K. MEES [J. S. M. P. E. 

mum occurs at approximately equal negative and print densities. 
Fig. 35 shows the modulation of the photoelectric cell current as a 
function of print density for the same series of negative densities as 
those in Fig. 34. This also shows that over a wide range of conditions 
the volume obtained is within 2 decibels of the maximum. The con- 
ditions for maximum modulation and minimum harmonic distortion 
are approximately at the same point. The optimal conditions may 
vary with the particular optical recording system used, since this 
would affect the amount of stray light scattered into the unexposure 
part of the track. The point of minimum distortion occurs when 
the lack of sharpness, due to light scattered by the optical system and 
by the photographic emulsion in the recording process, is compen- 
sated for by a corresponding spreading of the image in the printing 
process. As stated above, a change in the characteristics of the re- 
cording system may vary the amount of distortion in the negative 
sound record. It would, therefore, be necessary to change the print- 
ing conditions to obtain the maximum amount of compensation in the 
printing process. 

Improvements in the reproduction of sound by photographic means 
will depend, in the future as in the past, upon intensive scientific 
research in relation to sound, electricity, and photography. I be- 
lieve that this account of the work which has been done on the pho- 
tographic aspects of the subject will show that research in that field 
is being prosecuted with energy. The same is true of the work 
which is being done upon sound and upon electrical apparatus. We 
may expect, therefore, that the quality of the sound heard in the 
motion picture theaters will improve steadily, and that eventually it 
will leave nothing to be desired by the most critical audiences. 

REFERENCES 

1 MEES, C. E. K.: "The Photographic Reproduction of Tone," Phot. J., 64 
(1924), p. 311. 

2 CALLIER, A.: "The Absorption and Scatter of Light by Photographic 
Negatives, Measured by Means of the Martens Polarization Photometer," 
Phot.J., 49 (1909), p. 200. 

1 HALL, V. C.: "The Decibel in the Motion Picture Industry," /. Soc. Mot. 
Pict. Eng., XVIII (Mar., 1932), No. 3, p. 292. 

4 KtfsxER, A., AND SCHMIDT, R. : "Einfluss des Callier-Effekts auf die Wieder- 
gabe von Lichttonaufzeichnungen," Kinotechnik, 12 (1930), p. 602. 

* MILLER, D. C.: "Science of Musical Sounds," Macmillan Co., New York, 
1926, p. 92. 



April, 1935] SOUND RECORDING 325 

RENWICK, F. F.: "The Underexposure Period in Theory and Practice," 
Phot.J., 53 (1913), p. 127. 

BIBLIOGRAPHY 

SANDVIK, O., HALL, V. C., AND STREIFFERT, J. G.: "Wave-Form Analysis of 
Variable- Width Sound Records," /. Soc. Mot. Pict. Eng., XXI (Oct., 1933), No. 4, 
p. 323. 

HARDY, A. C.: "The Rendering of Tone Values in the Photographic Record- 
ing of Sound," Trans. Soc. Mot. Pict. Eng., XI (1927), No. 31, p. 475. 

COOK, E. D.: "The Aperture Effect," /. Soc. Mot. Pict. Eng., XIV (June, 
1930), No. 6, p. 650. 

FOSTER, D.: "The Effect of Exposure and Development on the Quality of 
Variable-Width Sound Recording," /. Soc. Mot. Pict. Eng., XVII (Nov., 1931), 
No. 5, p. 749. 

MAURER, J. A.: "The Photographic Treatment of Variable-Area Sound- 
Films," J. Soc. Mot. Pict. Eng., XIV (June, 1930), No. 6, p. 636. 

SANDVIK, O., AND HALL, V. C.: "Wave-Form Analysis of Variable-Density 
Sound Recording," /. Soc. Mot. Pict. Eng., XIX (Oct., 1932), No. 4, p. 346. 

MACKENZIE, D.: "Straight-Line and Toe Records with the Light- Valve," 
/. Soc. Mot. Pict. Eng., XVII (Aug., 1931), No. 2, p. 172. 

JONES, L. A.: "Photographic Reproduction of Tone," /. Opt. Soc. Amer., 5 
(May, 1921), No. 3, p. 232. 

TUTTLE, C., AND MCFARLANE, J. W.: "The Measurement of Density in 
Variable-Density Sound-Films," /. Soc. Mot. Pict. Eng., XV (Sept., 1930), No. 3, 
p. 345. 

NICHOLSON, R. F.: "The Processing of Variable-Density Sound Records," 
J. Soc. Mot. Pict. Eng., XV (Sept., 1930), No. 3, p. 374. 

SANDVIK, O.: "Apparatus for the Analysis of Photographic Sound Records," 
J. Soc. Mot. Pict. Eng., XV (Aug., 1930), No. 2, p. 201. 

MACKENZIE, D.: "Sound Recording with the Light-Valve," Trans. Soc. Mot. 
Pict. Eng., XH (1928), No. 35, p. 730. 

SANDVIK, O., HALL, V. C., AND GRIMWOOD, W. K. : "Further Investigations of 
Ground-Noise in Photographic Sound Records," J. Soc. Mot. Pict. Eng., XXII 
(Feb., 1934), No. 2, p. 83. 

SANDVIK, O.: "A Study of Ground-Noise in the Reproduction of Sound by 
Photographic Methods," Trans. Soc. Mot. Pict. Eng., XII (1928), No. 35, p. 790. 

CRABTREE, J. I., SANDVIK, O., AND IVES, C. E.: "The Surface Treatment of 
Sound-Film," J. Soc. Mot. Pict. Eng., XIV (Mar., 1930), No. 3, p. 275. 

JONES, L. A., AND SANDVIK, O.: "Photographic Characteristics of Sound 
Recording Film," J. Soc. Mot. Pict. Eng., XIV (Feb., 1930), No. 2, p. 180. 

DIMMICK, G. L., AND BELAR, H.: "Extension of Frequency Range of Film 
Recording and Reproduction," /. Soc. Mot. Pict. Eng., XIX (Nov., 1932), No. 5, 
p. 401. 

STRYKER, N. R.: "Scanning Losses in Reproduction," /. Soc. Mot. Pict. Eng., 
XV (Nov., 1930), No. 5, p. 610. 

COOK, E. D. : "The Aperture Alignment Effect," /. Soc. Mot. Pict. Eng., XXI 
(Nov., 1933), No. 5, p, 390, 



326 C. E. K. MEES 

SCHMIDT, R., AND KUSTER, A. : "Analysis of Sound Quality with the Variable- 
Density Recording Method from Sensitometric Data," /. Soc. Mot. Pict. Eng., 
XXI (Nov., 1933), No. 5, p. 374. 

KREUZER, B. : "Noise Reduction with Variable- Area Recording," /. Soc. Mot. 
Pict. Eng., XVI (June, 1931), No. 6, p. 671. 

SILENT, H. C., AND FRAYNE, J. G.: "Western Electric Noiseless Recording," 
/. Soc. Mot. Pict. Eng., XVIII (May, 1932), No. 5, p. 551. 

SHEA, T. E., HERRIOTT, W., AND GOEHNER, W. R.: "The Principles of the 
Light-Valve," /. Soc. Mot. Pict. Eng., XVIII (June, 1932), No. 6, p. 697. 

DIMMICK, G. L. : "High-Frequency Response from Variable-Width Records 
as Affected by Exposure and Development," J. Soc. Mot. Pict. Eng., XVII 
(Nov., 1931), No. 5, p. 766. 

SCHMIDT, R. : "On the Sensitometric Control of the Processing of Sound Films, 
Part I," Kinotechnik, 12 (1930), p. 574; Part II; KUSTER, A., AND SCHMIDT, R., 
ibid., 13 (1931), p. 123. 

FISCHER, F., AND LICHTE, H.: "Tonfilm aufnahme und Wiedergabe nach 
dem Klangfilm-Verfahren (Sound-Film Recording and Reproducing by the 
Klangfilm Process)," S. Hirzel, Leipzig, 1931. 

EGGERT, J., AND SCHMIDT, R.: "Einfuhrung in die Tonphotographie (Sound 
Photography)," S. Hirzel, Leipzig, 1932. 



CERTAIN PHASES OF STUDIO LIGHTING* 



C. S. WOODSIDE** 



Summary. Certain technical phases of studio lighting are discussed, including 
the new line of Mazda lamps for photographic purposes; new materials and contours 
for reflectors; an analysis of the several means of achieving diffusion; and the pro- 
jection of shadows upon backgrounds. 

During the course of preparation of a film for use by amateur 
cinema clubs we had occasion to study several phases of studio light- 
ing. Of vital importance to the amateur and of considerable interest 
to the professional is the quantity of photographically effective light 
that can be squeezed out of a single unit. The operation of lamps 
above their rated voltage was discussed in 1933 by Beggs and Palmer. 1 
Since that time a series of Mazda lamps has been designed to operate 
at very high temperatures and correspondingly high photographic 
efficiencies. Table I gives their important characteristics. 

TABLE I 

Characteristics of High-Temperature Incandescent Lamps 



Service 

No. 1 Photoflood 
No. 4 Photoflood 
Movieflood 
Movieflood 
Movieflood 
General* 



tWatts 

250 
1000 
1000 
1500 
2000 
1000 



Volts 


Bulb 


Base 


105-120 


A-21 


Medium Screw 


105-120 


PS-35 


Mogul Screw 


105-120 


PS-52 


Mogul Screw 


105-120 


PS-52 


Mogul Screw 


105-120 


PS-52 


Mogul Screw 


115 


PS-52 


Mogul Screw 



Light Center 
Length 





Finish 


fLumens 


fColor Temp. 


fLife Hours 


(iuche: 


No. 1 Photoflood 


I. F. 


8,875 


3450K. 


2 


3'A 


No. 4 Photoflood 


I.F. 


33,900 


3400K. 


10 


7 


Movieflood 


Clear 


32,500 


3360K. 


15 


9Vt 


Movieflood 


Clear 


49,090 


3370K. 


15 


9Vi 


Movieflood 


Clear 


69,200 


3430K. 


15 


9Vi 


General 


Clear 


20,500 


2990K. 


1000 


9Vt 



t Watts, lumens, color temperature, and life are given on the basis of opera- 
tion at 115 volts. 

t Shown for purposes of comparison. 



* Presented at the Spring, 1934, Meeting at Atlantic City, N. J. 
** Westinghouse Lamp Co., Bloomfield, N. J. 



327 



328 



C. S. WOODSIDE 



[J. S. M. P. E. 



The figures of Table I indicate the great increase of illumination 
that results from operating the lamps at high temperatures. The 
spectral quality of this high-temperature illumination is different 
from that of general service lamps. Fig. 1 illustrates the manner in 
which the radiation in the blue-green section of the spectrum in- 
creases with the temperature. This, of course, increases the photo- 



,ok ULTRA 

VIOLET 



VISIBLE LIGHT 




3500 K. 



INFRA RED 



3250 K. 



2900 K 



2840 K. 



2640 K. 



50 .60 .70 .80 

WAVE LENGTH IN MICRONS 

FIG. 1. Distribution of visible energy radiated from 
tungsten filament. 

graphic effectiveness of the light source, as indicated in Fig. 2. 
Whereas panchromatic film is only slightly more sensitive to light of a 
high color temperature, the gain with orthochromatic film is from 
25 to 50 per cent. 

As a practical example, a broadside unit employing two 1000- watt 
general service lamps could be relamped with two 1000- watt photo- 
flood lamps, with an increase in photographic effectiveness of 70 per 



April, 1935] 



STUDIO LIGHTING 



329 



cent for panchromatic film and 90 per cent for orthochromatic film. 
This increased effectiveness is attained with no increase of power or of 
radiant heat. Radiant heat is an important factor which may 
cause considerable discomfort on the set. An effective method of 
eliminating the radiant heat consists in placing a greenish blue glass 
cylinder over the lamp. The Macbeth Day lighting Company has 
made an exhaustive study of this subject, and is equipped to provide 
glasses of various heat-absorbing characteristics. Some of the glasses 
will remove 90 per cent of the radiant heat while absorbing only 20 
per cent of the light. The heat is of course re-radiated from the 
cylinder, and should be removed by a blower or air duct. The effect 




24OO 



3600 



2600 2600 3000 3200 34OO 
COLOR TEMPERATURE IN DEGREES KELVIN 

FIG. 2. Relation of photographic effectiveness vs. color 
temperature of incandescent tungsten. 

upon the set, however, is a complete absence of discomfort from heat. 
Reflecting Equipment. -Recent developments in reflecting equip- 
ment have included both contour changes and new materials. In 
addition to the customary silver, chromium, nickel, and porcelain 
enamel, we now find rhodium plate, and the Alzak processed alumi- 
num. Rhodium is nontarnishable, and the plating process involved 
permits contours of extreme accuracy. The Alzak processed alumi- 
num, fabricated by a new electrolytic process, has a very high reflec- 
tion factor and is available in polished or matte finish. Manufactur- 
ers will undoubtedly produce some effective light-weight equipment 
using this material. 



330 C. S. WOODSIDE [J. S. M. P. E. 

TABLE II 

Reflection Factors of Commonly Used Materials 

Material Surface Reflection Factor 

Silvered Glass Specular 0.90 

Alzak Aluminum Specular 0.83 

Alzak Aluminum Matte 0.85 

Rhodium Specular 0.75 

Chromium Specular . 64 

Nickel Specular 0.64 

Aluminum Matte 0.65 

Porcelain Enamel Semi-matte 0.75 

Table II gives the reflection factors of the materials commonly 
used in projection. Specular reflection is important in equipment 
designed to produce high-intensity beams and highlights. Diffuse 
reflection is important in broadside units for general illumination. 

The usual reflector contours employed in the past have been spheri- 
cal or paraboloidal. The newest equipment for spotlighting employs 
an ellipsoidal contour. Since this subject forms the basis of another 
paper we will not discuss it here. 

Diffusion. Inside frosted lamps combined with specular re- 
flectors produce smooth beams of moderate concentration. Where a 
more highly concentrated beam is required it is necessary to use a 
clear lamp with a concentrated filament. Even with the biplane fila- 
ment lamps, the source of which is nearly solid, the projected beam 
reveals filament striations. To remove the striations a certain 
amount of diffusion is needed. The two general methods of attain- 
ing diffusion are, first, to change the reflector characteristics by 
stippling, faceting, or employing a series of concentric rings; and, 
second, by placing a diffusing screen in front of the housing. By the 
first method the beam is produced by a series of paraboloidal surfaces 
and a series of surfaces slightly off the theoretical curve. The beam 
intensity is decreased and the spread increased. Table III shows 

TABLE in 

Characteristics of Deep Parabolic Reflectors 

Surface Characteristic Beam Intensity Beam Spread 

(per cent) (inches) 

Smooth 100 26 

Concentric Rings 45 62 

Stippled 21 72 



April, 1935] 



STUDIO LIGHTING 



331 



results produced by deep parabolic reflectors of identical dimensions 
and material. The loss of beam intensity is serious unless a number 
of projectors are to be used, when the overlapping beams will produce 
approximately the same average intensity as the same number of 
smooth-surfaced reflectors. 

The second method has been used in many forms by both amateur 
and professional. Silk, tracing-cloth, figured glass, spread lenses, 
wire netting, and other materials have found employment. In an 
effort to find a suitable screen that would have the minimal effect upon 
the beam and yet display sufficient diffusing power to avoid objec- 
tionable streaks, we came across several materials that proved satis- 
factory. The new ellipsoidal reflector spotlight with objective lens 
produces a uniform spot of adjustable size. The older type of spot- 




FIG. 3. Modification of spot by various diffusers: (A) no diffuser; (B) 2- 
degree circular spread lens; (C) Luminex glass; (D) Velvex glass. 



light, employing a spherical mirror and condenser lens, is limited in 
focusing positions. Drawing the spot to a sharp focus produces fila- 



TABLE IV 



Diffusion Test: 



2000- W., 115-V. Biplane Lamp; 8 -Inch Condenser in 
Spotlight 17 Feet from Screen 



Diffusing Medium 
None 

2 Circular Lens 
2 Horizontal Spread Lens 
Bubble Glass 
Luminex Wired Glass 
Velvex Wired Glass 
Silk 



Foot-Candles 
on Screen 

450 

380 

310 

'320 

225 

215 

13 



Width of Beam 
(inches) 

16 
22 
29 

25 

32 

33 

156 



332 



C. S. WOODSIDE 



ment striations and chromatic aberrations. Any one of several 
diffusive media will correct this condition. Fig. 3 shows the change of 
appearance of a spot when such a diffuser is used. Under ordinary 
conditions the lamp is thrown out of focus to produce a smoother spot. 
A test made under these conditions with a 2000- watt 115-volt biplane 
filament lamp in a spotlight with an 8-inch condenser located 17 feet 
from the screen produced the results shown in Table IV. 

The 2-degree spread lens is not available in sizes greater than 9 
inches in diameter, and hence can not be used with the larger size 
floodlights and spotlights. With this equipment the wired sheet 

glasses, such as Velvex or Lumi- 
nex, are recommended, since 
the danger from broken glass 
is eliminated. When using high- 
wattage lamps it may be neces- 
sary to split the glass into strips 
to provide for expansion. The 
heat-absorbing glass cylinder 
previously mentioned will pro- 
vide some diffusion and conse- 
quently serve a double purpose. 
Projection of Shadows. It is 
frequently required to project 
a shadow upon a background. 
A point-source such as a bare 
lamp will produce a sharply de- 
fined shadow, but the wattage 
required to produce sufficient 
intensity is usually prohibitive. 
A more satisfactory method consists in using a focusing type of spot- 
light in which the lamp is located as far in front of the focus as pos- 
sible. vSuch an arrangement acts virtually as a point source, affording 
good definition and at the same time sufficiently high intensity. It 
is, of course, necessary to shield the background from the foreground 
lighting so that the shadow is not blotted out. This can be done by 
using overhead lights of the concentrating type, as shown in Fig. 4. 

" REFERENCE 

1 PALMER, M. W., AND BEGGS, E. W.: "Professional Motion Picture Photog- 
raphy with High-Intensity Short-Life Incandescent Lamps," /. Soc. Mot. Pict. 
Eng., XXI (Aug., 1933), No. 2, p. 126. 




HISTORICAL NOTES ON X-RAY CINEMATOGRAPHY* 
R. F. MITCHELL** AND L. G. COLEf 

Summary. The history of the various attempts to make x-ray movies is briefly 
traced. In the beginning there was only visual examination. When it became pos- 
sible to take still x-ray photographs, the next step was to "animate" them by producing 
a series of stills showing different phases of movement. 

With improvements in x-ray apparatus and reinforcement screens the time of ex- 
posure was reduced so that it became possible to take a series of x-ray photographs on 
a long strip of film. Speeds up to four pictures per second have been reached, though 
one experimenter claims speeds up to fourteen frames a second. 

The "frames" so taken are copied on 35-mm. or 16-mm. film and shown in regu- 
lar equipment. The 16-mm. display device described attracted great interest at the 
Chicago Century of Progress Fair and has since excited like interest at other popular 
expositions. 

The methods employed to animate x-ray films are not only of historical importance, 
but were of considerable legal importance in connection with patent litigation on 
regular motion picture animation processes. 

The subject of this paper was originally prompted by the great 
interest aroused by the display device illustrated in Fig. 4 
exhibited at the Century of Progress Fair at Chicago, 111., 
1934. The historical background has been worked out in what is 
believed to be a comprehensive and impartial manner, although 
no claim is made as to its legal accuracy. It is felt that this survey 
is of historical value for the records of the Society. 

In 1896, Hammeter of Baltimore and Wolf Becker in Europe, made 
attempts to use opaque solutions to render visible the lumen of the 
hollow viscera. 

On July 24, 1897, before La Societ de Biologie, Roux and Baltha- 
zard 1 presented their first reports of the use of opaque bismuth mealt 
in the study of the motor function of the stomach. Their full report 
was published in 1898. They seem to have been the first to conceive 
the possibility of x-ray cinematography of the gastro-intestinal tract, 

* Presented at the Fall, 1934, meeting at New York, N. Y. 
** Bell & Howell Company, Chicago, 111. 
t Cole Laboratories, New York, N. Y. 
% A paste of water and bismuth sulfate, 



334 R. F. MITCHELL AND L. G. COLE [J. S. M. p. E. 

and the first to accomplish it with frogs. Their efforts merit consider- 
able prominence in view of the time it took for x-ray motion pictures, 
as such, to become an accomplished fact. 

H. P. Bowditch, Professor of Physiology at Harvard University, 
suggested to W. B. Cannon, then a medical student, that the x-rays 
be used as a means of studying deglutition under normal conditions. 
Cannon, in 1896, used free bismuth in studying the stomach of a 
cat. F. H. Williams, of Boston, deserves a good deal of credit for his 
efforts in this type of work. Williams, assisted by Cannon, on Sep- 
tember 23, 1899, performed further experiments 2 of this type. 

Then for almost half a decade there was a silent era broken only by 
the contributions of O. Kraus and Lommel. 

In 1903-04, as a result of gradual improvements in apparatus, the 
time of exposure had been reduced to fifteen or twenty seconds, so 
that roentgenograms could be made of the kidney stone while its 
motion due to respiration was stopped. Sufficiently rapid exposure 
to avoid the motion of gastric peristalis, however, continued to be 
impossible. Roentgenographs of the stomach were so blurred that 
they were of little or no diagnostic value. Methods of exciting the 
x-ray tube, however, were such as to enable one to observe fluoro- 
scopically the size, shape, and position of the stomach, as well as 
some of the grosser lesions of the stomach, when present. 

In the United States, the resurrection and further development of 
gastro-intestinal roentgenology dates from 1905. In that year, 
Hulst, who had visited Rieder in Munich, informally presented 
roentgenographs of the gastro-intestinal tract at a meeting of the 
American Roentgen Ray Society in Baltimore. The following year 
he presented an illustrated comprehensive paper on the roentgeno- 
graphic method of examination of the gastro-intestinal tract 
before the American Roentgen Ray Society, which was so complete 
that it furnished a new impetus to the method Williams had sug- 
gested seven years previously. At the same meeting (American 
Roentgen Ray Society, 1906) Snook 3 first presented the electrical 
facts upon which he developed and made practicable the transformer 
that superseded the static machine and coil. Snook developed the 
transformer in 1906, and this, with the improvement and application 
of the intensifying screen, enabled exposures to be made in a frac- 
tion of a second : thus the blur due to movement of the stomach was 
eliminated. 

Medical men differed concerning the efficiency of diagnosis from a 



April, 1935] X-RAY CINEMATOGRAPHY 335 

single plate as compared with fluoroscopic examination, because the 
various roentgenograms of the same stomach differed so much in 
appearance that a conclusion could not be drawn from the evidence on 
a single plate. The proponents of the morphologic basis of x-ray 
diagnosis, particularly Cole 4 and his associates, proceeded to obviate 
this objection by making a series of plates in as rapid succession as 
possible. This method was called ' 'serial roentgenography . ' ' 

During the years 1905 to 1909, fluoroscopic exploration of the gastro- 
intestinal tract became very popular on the European continent. 

The experiments of Roux and Balthazard were resumed by Levy- 
Dorn in 1905, and in 1907 by A. Kohler. These men took a large 
number of radiograms at a maximum rate of about one per second, 
and then combined the images in a series. 

Levy-Dorn, for instance, radiographed 20 to 22 phases of movement 
in 20 seconds, and each phase was repeated twice, one after another. 

In 1908, Eijkmann succeeded in obtaining x-ray cinematographs of 
the act of deglutition. The slight thickness of the organs of the throat 
had enabled him to take instantaneous radiographic poses, that is, 
poses corresponding to an opening of the primary current of the 
Ruhmkorff coil. Unfortunately, his process, which was easy to carry 
out with suitable electrical controls and very small plates, laid the 
operator open to all the dangers connected with the excessive use of 
x-rays, causing the hair to fall and provoking erythema. In addition, 
it could not be used for large organs such as the stomach and the 
heart. This would have necessitated enormous intensities of x-rays 
for radiograms of such short duration, and such intensities were not 
available in those days. 

But great improvements soon began to be made in radiological 
apparatus, which, with the elimination of the closing currents, gave a 
very high intensity ray, namely, 100 or more milliamperes. 

In 1910, at the same Congress where Groedel had presented his 
roentgenocinematographs of the heart, Biesaldski and A. Kohler illus- 
trated a process, worked out by Kohler during the previous year, 
consisting in the cinematographic photography of moving images 
as they appear upon the fluorescent screen. This brilliant idea 
occurred again, later on, in 1911, to two French radiologists, Lomon 
and Comandon. Their technic, which was used for the first time in 
the Physics Laboratory of Professor Broca of the Faculty of Medicine 
of Paris, also consists in cinematographic photography, with a speciaj 
lens, of the images on the fluorescent screen, 



336 



R. F. MITCHELL AND L. G. COLE 



[J. S. M. P. E. 



Until 1909, Austria and Germany led the world in this work; 
among the leaders were Holzknecht, Strauss, Rieder, Schwarz, 
Kreuzfuchs, Groedel, Albers-Schonberg, Haenisch, and Kienbock. 

In 1910 Cole and his associates 5 developed the idea, and regularly 
made a series of x-ray pictures to which they gave the name "serial 
roentgenography." To these roentgenograms some 10 or 12 pictures 
were made at intervals from 4 to 10 seconds apart, so that in this 
series all the phases of the gastric motor phenomenon were depicted. 

In 1909, Kaestle, Rieder, and Rosenthal 6 made roentgenocine- 
matographic films of the stomach under the term "bioroentgenog- 

raphy." They assembled roent- 
genograms of a normal individ- 
ual made on 18 X 24-cm. plates, 
exposure 20 to 22 sec., 30 to 60 
milliamperes at 220 volts. These 
were reproduced cinematographi- 
cally to illustrate the normal 
motor phenomena of the stomach. 
One of the most notable im- 
provements was the development 
of the reinforcement screen by 
Gehler of Leipzig. The reinforce- 
ment screen is formed of a plate 
that is very easily penetrated by 
the x-rays; it is covered on one 
side by a salt that transforms the 

rays into ordinary light. If this 
FIG. 1. Drawing of roentgenocine- . . , . 

matographic apparatus. covered side is closely attached to 

the sensitive surface of the photo- 
graphic plate, it is possible to make much shorter exposures than 
would be necessary otherwise, using the same intensity of ray, with 
excellent results. It was thus possible to reduce considerably the 
time required to make a good radiograph. 

In 1909 Groedel, making use of radiographic films placed between 
two reinforcement screens, instead of photographic glass plates, 
was able to obtain, at 60 to 70 cm., radiograms of the thorax and 
abdomen in one-fifth to one-twentieth of a second. By means of 
this improvement and of a special mechanism for changing the plates, 
Groedel made a film of the movements of the heart, which he obtained 
by a series of 6 radiograms per second. This he presented before 




April, 1935] X-RAY CINEMATOGRAPHY 337 

the Congress of Internal Medicine, held in that year at Berlin. Some 
question has been raised, however, as to whether those films were 
true x-ray motion pictures or an assembly of a series of stills. 
Groedel's use of film instead of plates and the use of two reinforce- 
ment screens was later improved on by Luboshez, who was instru- 




FIG. 2. Roentgenocinematographic apparatus de- 
signed to move film 10 inches wide: installed in the 
Joseph Purcell Memorial Laboratory at the Fifth Ave- 
nue Hospital, New York, N. Y. ; apparatus opened for 
threading. 

mental in promoting the commercial development of the double- 
emulsion film used between two reinforcement screens, a system 
still in use today. The invention of the method is ascribed by Fuchs 
to M. Levy in 1897. 7 
An x-ray motion picture machine developed by Cole was demon- 



338 R. F. MITCHELL AND L. G. COLE [J. S. M. P. E. 

strated at Detroit, Oct., 1910. It moved a film 8 inches wide, and 
had a relatively narrow gate, 5 inches. Further experiments with 
x-ray cinematography on this basis were conducted by Cole 8 with the 
apparatus shown in Fig. 1, and conclusive results regularly obtained 
by 1913. The equipment was further improved to the form of the 
elaborate apparatus shown in Figs. 2 and 3, now in use at the Pur- 
cell Memorial Laboratory, Fifth Avenue Hospital, New York, N. Y. 
This latest machine employs the principle of the regular motion 
picture machine. It uses film 10 inches wide, perforated in a special 
perforating machine, in exactly the same manner in which a standard 
motion picture camera moves a smaller film. With this apparatus 
we are able to make true motion pictures of the stomach at speeds 
up to four "frames" per second. The unused film is contained in a 
magazine, and is shown under the near end of the table. After pass- 
ing around wheels with sprockets the film is threaded between intensi- 
fying screens and then through rollers back into another magazine. 
In this photograph the apparatus is opened for threading. A worm- 
gear at the far end of the table enables it to be used in either the hori- 
zontal or vertical position, or at any desired angle. The same re- 
flecting mirror that was employed on the serial tables enables one 
to observe the action of the stomach both prior to making the film 
and during the time that the film is actually being exposed. 

Fig. 3 shows the roentgenocinematographic apparatus closed and 
ready for action. The tube is enclosed in a ray-proof box. The 
timing of the exposures is accomplished by a switch at the top of the 
tube stand, which may be used to break either the secondary or the 
primary current. The motor that drives the mechanical parts is ob- 
served in the foreground ; and just behind it is a speed-changing de- 
vice, similar in size and shape to the gear-shift of an automobile, 
which enables us to make a continuous roentgenographic film at 
various speeds. The film is exposed in the special x-ray "camera" 
which has a gate 10 inches long and 9 inches wide, allowing 1 / 2 inch 
of film on each side for the perforations. The film is developed in 
large tanks somewhat after the manner of developing film from aerial 
mapping cameras. Each individual "frame" is then copied on one 
frame of 35-mm. or 16-mm. film for showing in a regular projector. 

A wide roll of film was used by H. E. Ruggles 9 for x-ray cinematog- 
raphy of the heart. An account of the heart study made by 
Ruggles was presented by Chamberlain and Dock, 10 who describe the 
frames as 8 by 10 inches in size, and taken at the rate of 15 a second. 



April, 1935] 



X-RAY CINEMATOGRAPHY 



339 



Lomon and Comandon were able to obtain instantaneous photo- 
graphs on ultra-sensitive films, illuminating the fluorescent screen 
by means of very powerful beams of x-rays (corresponding to 300 ma. 
of a Coolidge therapy ampule on 80 kv.), using specially made 
fluorescent screens, and employing an //1. 55 lens (designed by C. E. 
Florlian, then engineer to the firm of Lacour-Berthiot) . The lens, 
which is the essential part of the apparatus, was calculated for 




FIG. 3. Another view of Fig. 2; apparatus closed and 
ready for use. 

photographing a flat surface emitting ultra-violet rays, and was 
therefore composed of substances that were transparent to them, 
namely, quartz, uviol glass, and glycerine. The cinematograph ap- 
paratus of the Pathe* firm was so modified as to increase as much as 
possible the time of the.exposure in relation to the time of obturation. 
The fluorescent screen and the cinematograph apparatus were made 
solid and firmly fixed on a rigid frame, which in its turn was fixed to 



340 R. F. MITCHELL AND L. G. COLE [j. s. M. P. E. 

the ground. The focusing was very difficult on account of the 
slight depth of the field of the lens, and once determined accurately, 
was fixed permanently. Among the various fluorescent screens ex- 
amined, Lomon, after many trials, chose those of calcium tungstate, 
like those used for reinforced screens, but coarse-grained. 

Three seconds is sufficient time in which to obtain at least two 
revolutions of the heart, and therefore that period was not exceeded 
for cinematographing them. The method has not been widely used 
on account of the danger either to the apparatus or to the patient 
(x-ray burns) . 

Direct cinematography of the fluoroscopic screen was attempted by 
Caldwell in New York in 1910 and by Kay in England, who did ex- 
tensive work in this connection. 

B. E. Luboshez, at the Italian Congress of Medical Radiology de- 
scribed in 1928 a system of direct photography in the center of the 
fluorescent screen. His first apparatus differed from the regular type 
by the use of a special lens having the exceptional speed of //0.625, 
invented by himself. 

In 1930, Busi 11 described before the Medical Society of Bologna a 
method of taking one-quarter of a radiographic frame (24 X 30 cm.) 
of a double-emulsion photographic plate between reinforcement plate 
and screen. However, the method had been widely employed in 
this country for a considerable period prior to 1930. 

R. Janker, of Boon, described an apparatus in 1931 in which the 
film, turning rapidly, passes between two strips of reinforcement 
screen, which also run with the same speed and the same stops as the 
film, but in the opposite direction; so that the new film is continually 
being placed between fresh and rested sections of the reinforcement 
screen, that is to say, sections in which the fluorescent light is already 
extinguished. Janker used his method with excellent results, but, it is 
believed, only on animals: the danger of lesions to man is still too 
serious. 

A newer apparatus shown in 1932 by Luboshez at the National 
Congress of Medical Radiology at Parma, has, among other improve- 
ments, four object glasses or lenses, instead of one. The apparatus 
demonstrated at the International Congress of Educational and 
Teaching Cinematography, April, 1934, uses 16-mm. films and takes 
pictures at normal as well as at an accelerated speed. It works 
electrically and automatically with the cardiological apparatus; it 
has a device for counting the meterage for a film 30 meters long and 



April, 1935] X-RAY CINEMATOGRAPHY 341 

for a duration of five minutes per reel. Finally, it is complete for 
every form of photographic manipulation, and can be loaded or un- 
loaded in the light. For a normal picture exposure of several sec- 
onds it is sufficient to feed the Coolidge type ampule with 50 ma. at 
75 kv. : a modest regimen representing the greatest safety for the 
apparatus and the patient. 

We have now reached the logical point where we can discuss the 




FIG. 4. Continuous display apparatus, 
exhibited at the Century of Progress Exposi- 
tion, 1934, at Chicago; the mirrors and path 
of the light rays are shown in the left-hand 



film shown at the^ World's Fair at Chicago in connection with the 
"Blonde Lady in Black"; the display is illustrated in Fig. 4. 
The arrangement involves a continuous projector with a small mirror 
placed just ahead of the projection lens, the pictures being reflected 
to another mirror located directly behind a transluscent screen in 
the position shown. When the projector is switched off, little or 
no trace of the screen is evident. However, as soon as the button 



342 R. F. MITCHELL AND L. G. COLE [J. S. M. P. E. 

on top of the box is pressed, a picture of very satisfactory 
brilliance is visible through the thin black material employed. To 
the layman the model is interesting as illustrating the size and shape 
of the stomach and its movement when digesting foods. To the 
medical student it illustrates accurately the several distinct motions 
that constitute the complex gastric motor phenomena. To the mo- 
tion picture engineer it is interesting because of the manner in which 
the various scenes were made. 

The first sequence of the film to be shown was made by assem- 
bling and copying x-ray plates that had been taken at various intervals 
in the order of progress of the cycle. Early motion picture films il- 
lustrating the movement of the stomach had been made in that 
manner. This crude cinematographic presentation of what was called 
serial roentgenograms established that the then-prevalent idea that 
the stomach held food for several hours until it was digested and 
then rapidly evacuated it was erroneous. 

The first motion picture films that we made were made with an 
early motion picture projector used as a camera, installed in the 
wall of a dark room. The regular projection lens was used to photo- 
graph a series of x-ray plates assembled in the proper order. The 
film in the projector was moved past the gate by hand as each suc- 
ceeding phase was placed before the illuminating box. The x-ray 
films were registered by arranging them according to the progress of 
the peristaltic wave, placing one plate upon another, and making two 
dots on each plate. The dots were then registered by hand on a 
given spot on the illuminating box. This procedure of making 
cinematographic films of x-ray plates did not exactly constitute ani- 
mation, although it was very close to it. 

In one case, however, prints were made of a set of x-ray films, and 
cut-outs were made of the outline of the stomach, showing accurately 
its size and shape, and the peristaltic waves. The prints were then 
reversed, so that the plain white cardboard back of the cut-out was 
photographed, instead of the picture on the cut-out. These were 
photographed on a film as described before and the film was shown in 
Detroit in the early part of October, 1910. This procedure of photo- 
graphing the white cardboard cut-out of the stomach probably was 
the first attempt at animation, and this film, therefore, is of interest 
from a historical standpoint as it has already become of interest from a 
legal standpoint, in that it antedated, by approximately a decade, the 
animation of Felix the Cat by Pat Sullivan. This early animation of 



April, 1935] X-RAY CINEMATOGRAPHY 343 

x-rays has been of great significance in connection with patent litiga- 
tion covering animation of regular motion picture films. 

The second section is an animated film, consisting of 88 line-draw- 
ings made by Cole without a peg-board and without intermediates. 
Beginning at one end of the cycle and going right through the 88 
drawings, there is only a slight backlash on the greater curvature in 
one short period of the cycle. These drawings were based on an in- 
tensive study of innumerable sets of roentgenograms showing the 
complexity of the gastric motor phenomenon. 

The line-drawings were filled in with black by Carpenter and 
Goldman, and photographed by them. The scene was the final scene 
of a two-reel animated film, illustrating and analyzing the complexity 
of the gastric motor phenomenon, and showing the various types of 
stomach. This film, entitled The Complex Gastric Motor Phenomenon, 
was shown at Chicago in 1923. 

The third section is a more accurate presentation of advanced 
studies made by Cole in the late twenties. Having learned the tricks 
and science of animation from Carpenter and Goldman, these were 
applied to a more accurate presentation of the advanced studies 
that had been made of the gastric motor phenomenon. This film 
also shows more accurately the altered shape of the cap to conform 
to the stages of pyloric systole (contraction of the pyloric end of the 
stomach). 

The fourth section is a true roentgenocinematographic film, made of 
various phases of the same cycle, and was first shown at the Third 
Radiological International Congress at Paris, in 193 1 . In proof of the 
fact that this is a true roentgenocinematographic film, the rapid move- 
ment of the small intestine is shown on the same film showing the 
slow movement of the stomach, proof positive that these are not as- 
sembled films of various cycles. The modern machine with which 
this film was made is shown in Figs. 2 and 3. 

REFERENCES 

1 COLE, L. G., AND COLLABORATORS: "The History of the Radiological Ex- 
ploration of the Mucosa of the Gastro-Intestinal Tract," Bruce Pub. Co. 
(New York), 1934. 

* Williams, F. H.: "Roentgen Rays in Medicine and Surgery," MacMillan 
Pub. Co. (New York), 1901. 

Snook, C.: Trans. Amer. Roengten Ray Soc., 1906. 

4 COLE, L. G., AND COLLABORATORS: Archives. Amer. Roentgen Ray Soc., 1911. 

6 COLE, L. G., AND COLLABORATORS: Amer. Quarterly of Roentgenology, No 1. 
(March, 1912). 



344 R. F. MITCHELL AND L. G. COLE [J. S. M. P. E. 

6 KAESTLE, RIEDER, AND ROSENTHAL: "The Bioroentgenography of the In- 
ternal Organs," Archives Amer. Roentgen Ray Soc., No. 119 (June, 1910). 

7 JARRE, H. A.: "Science of Radiology," C. C. Thomas (Springfield, 111.), 1933, 
p. 100. 

8 COLE, L. G.: Amer. J. of Roentgenology (March, 1914). 

9 RUGGLES, H. E.: Radiology, 5 (1925), p. 444. 

10 CHAMBERLAIN, W. E., AND DOCK, W.: Radiology, 7 (1926), p. 185. 

11 Busi, A.: "X-Ray Cinematography," Internal. Rev. Ed. Cinemat. No. 2, 
(Feb., 1934), p. 84. 

DISCUSSION 

DR. COLE: This motion picture projector is perhaps of historical interest in 
indicating the crude way in which this work was done. I obtained it about 1906, 
but I am not sure of the date. We set the camera into the wall of the dark room. 
We turned the crank steadily by hand and moved the film along; then we set up 
the next one, and so obtained our first animated movies. There were only 20 
cut-outs, I believe, and they were repeated over and over again. 

Referring to Figs. 2 and 3, the table may work either horizontally or vertically, 
so that the patient may be x-rayed lying on his abdomen or in the erect posture 
showing the change in the position of the stomach and the relation of the viscera 
to each other. The principle of the machine is essentially the same as that of the 
motion picture camera. We used the Geneva movement. First the film goes 
through the typical sprockets with the Geneva movement, and comes up over the 
rollers just beneath the bed of the table, and out, where it is picked up by the 
second line of Geneva movements which shift it and stop it. It is taken up in a 
casette or box at the other end. Each time the film moves (it must come to a 
standstill, of course) it must be registered. Then the intensifying films must be 
separated the suction must be overcome. We even use compressed air to aid in 
separating the film. 

The gear box, about the same in size, shape, and characteristics as the gear- 
shift of an automobile, has four or five shifts enabling us to take an exposure every 
two seconds, every second; one a second, two a second, three a second and four a 
second. 

One of the greatest difficulties that we had to overcome was registration. It 
was accomplished by means of an accessory punch. Two punches perforate the 
film accurately and the film is then positioned so that the punch holes fit firmly 
over tapered pegs. 

MR. CRABTREE : What is the width of the film ? 

DR. COLE: Ten inches. It is registered approximately to a thousandth of an 
inch. 

MR. CRABTREE : What is the advantage of this scheme over photographing the 
image on a fluorescent screen? 

DR. COLE : The stomach just can't be photographed on the fluoroscopic screen; 
or at least I won't say it can't be done, but it hasn't been done as yet. The 
detail obtained by such photographing of the screen is not sufficient for diagnosis, 
even of the heart and lungs, and detail is the essential factor in the diagnosis of 
both gastro-intestinal and lung lesions, the heart less so. 



April, 1935 J X-RAY CINEMATOGRAPHY 345 

MR. CRABTREE: What has been the advantage of the motion picture as 
compared with the still, from a diagnostic standpoint? 

DR. COLE: From organic lesions probably nothing. It has, however, been 
very essential in proving the accuracy of the complex motion of the stomach. 
The action of the stomach is made up of seven motions, which we had recognized 
in the still pictures; enough stills had been taken at random and set up so 
that we could study them. The ultrascientific man said, "Yes, but that isn't a 
true motion picture. You haven't proved a systole and diastole of the pyloric 
end of the stomach." 

I wanted to prove that these contentions, as far as the physiology and motor 
phenomena of the stomach were concerned, were correct. It would be of value 
in the study of nervous indigestion, functional derangements of the stomach, and 
so forth. But in the prosperous days, we didn't have functional derangements of 
the stomach; and now that we don't have prosperous days, we don't have money 
enough to buy the films. A roll of film like this costs about $50, and we don't 
get $50 for an examination, especially for nervous indigestion. We might get it 
for cancer, but it isn't necessary for that. 

MR. CRABTREE: How close to the danger limit are you going in exposure? 
How long an exposure can the patient withstand with safety? 

DR. COLE: I believe he could stand at least four cycles of 20 seconds each a 
minute and twenty seconds provided we had filters. It must also be understood 
that the exposure is not continuous. It is intermittent, whereas with the photo- 
graphing of the screen, the x-ray is going through the patient continuously. With 
this method, it is only a flash and then you have to have the period for moving the 
film. 



SYMPOSIUM ON CONSTRUCTION MATERIALS FOR 
MOTION PICTURE EQUIPMENT* 

APPLICATIONS OF STAINLESS STEELS IN THE MOTION PICTURE 

INDUSTRY 

W. M. MITCHELL** 

Summary. The development of the stainless steels is related with the important 
position they have assumed in chemical and other industries because of corrosion 
resistance and unique properties. The steels suitable for use in motion picture in- 
dustries are classified into two groups: steels for structural purposes, and steels for 
resistance to corrosion. A brief description of these steels and their possible applica- 
tions in the industry are given. 

The stainless steels are a group of ferrous alloys notable for high 
resistance to corrosive attack and for freedom from tarnishing when 
exposed to the atmosphere. Their commercial prominence began 
about 1914 with the introduction by H. Brearley, of the Brown- 
Firth Laboratories, Sheffield, England, of cutlery made from stainless 
steel ; so-called because it did not discolor when in contact with acids 
of fruits and vegetables, as did the cutlery previously used. For 
some years applications were restricted to cutlery and cutting tools 
of various kinds, due, to a certain extent, to the inherent hardness of 
the alloy and the difficulty of producing it in various forms. But a 
softer modification, with equal corrosion resistance, was introduced 
about 1920. This was followed a few years later by the extensive 
use by one of the larger chemical concerns of the softer stainless steel 
for the construction of a plant for the manufacture of nitric acid. 

At about the same time iron-chromium-nickel stainless steels, 
which had been developed some years earlier by Strauss of the Krupp 
Gun Works, were intoduced from Germany. These possessed some- 
what higher corrosion resistance, were more easily workable, and at 
once came into popular use. Since that time development has 
progressed with great rapidity, and today there is scarcely an indus- 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 
'* Metallurgical Engineer, Subsidiary Manufacturing Companies, United 
States Steel Corp., New York, N. Y. 

346 



SYMPOSIUM ON MATERIAL FOR EQUIPMENT 347 

try that has not benefited in some way through the introduction of 
this group of metals. Chemical industries have received perhaps the 
greatest benefit, since here corrosion and its toll are greatest. But 
equally important applications have been found in the food-handling 
industries and for architectural ornamental work; and very recently, 
stainless steel and the light-weight construction made possible by it 
give promise of reduced operating costs of equipment for railroads and 
other agencies of transportation. 

Corrosion, the destroyer of metals, has always been with us. But 
until comparatively recent years its destructiveness was accepted as 
unavoidable, and the losses in various industries were absorbed as a 
part of the final cost of the manufactured product. This has natur- 
ally stimulated research and investigation, with the result that many 
new metals and alloys have been developed and brought into com- 
mercial use, affording in many cases considerable saving and economy. 
The most important of these are the group of stainless alloys or 
steels which depend for their corrosion resistance upon the consider- 
able proportion of the metal chromium that enters into their com- 
position. 

Since the invention of stainless steel by Brearley and the iron- 
chromium-nickel steels by Strauss, development has been rapid. 
Various modifications in the original compositions have been made to 
produce alloys or steels more suited for particular purposes, so that 
there are now some 50 varieties of what we are pleased to call "stain- 
less steel" in more or less regular production. Whether or not they 
should all be classed as "steels" may be open to question, since not 
all of them are hardenable on sudden cooling from a high temperature 
and, incidentally, the exact definition of the word "steel" is still a 
subject of discussion among metallurgists. However, they are all 
alloys of iron, and comply in general with the mechanical and metal- 
lurgical characteristics of what we mean by steel; hence the use of 
the term "stainless steel" for any or all of this group of alloys. 

The outstanding characteristic of this group is their high resistance 
to corrosive attack; and with this are combined excellent mechanical 
properties of strength, ductility or hardness depending, obviously, 
upon the composition of the particular steel. Varying with the com- 
position, we have steels ranging from those capable of great hardness 
and wear resistance to steels that have virtually no hardening ca- 
pacity whatever and are suitable for cold forming, bending, and simi- 
lar operations. 



348 W. M. MITCHELL [J. S. M. p. E. 

For applications in the motion picture industry the stainless steels 
can be divided into two groups : those suitable for structural applica- 
tions where strength, wear resistance, or hardness is required; and 
those for applications where resistance to corrosive attack is essential. 
It will be appreciated that these groups are not necessarily mutually 
exclusive and that certain steels find application for both structural 
and corrosion resisting purposes. 

Consider first applications for structural purposes. By structural 
we understand those applications where the mechanical properties of 
strength, hardness, wear resistance, etc., are desired. Some of them 
may be used for parts of machinery that must be maintained bright 
and free from rusting and at the same time possess the strength to 
withstand strains and sudden stresses and to be wear-resisting for a 
long life without frequent replacement that is, for shafts, levers, 
screws, gear-wheels, ratchets, etc., and the various parts that make up 
the mechanism of the modern motion picture camera. Those mem- 
bers of the stainless steel group that seem particularly suitable for 
this purpose are the following: 

(1) The original stainless cutlery steel of Brearley. Its composi- 
tion is approximately 12 per cent chromium, 0.35 per cent carbon, 
with the balance essentially iron. This is a true steel in that it may 
be made hard by sudden cooling from a high temperature. It is a 
very hard, strong, wear resisting material, suitable for parts that 
must withstand much wear and abrasion. It may be machined with 
reasonable success in the softened (annealed) condition but must be 
heat-treated (hardened) to develop proper hardness, strength, and 
corrosion resistance. As with all stainless alloys its surface must be 
properly polished for maximum resistance to rusting and tarnish. 
Table I gives average physical properties. 

TABLE I 
Properties of 12% Cr, 0.35% C Stainless Steel 

Heat Treated Annealed 

Ultimate Strength (Ib./sq. in.) 252,000 99,950 

Yield Strength (Ib./sq. in.) 210,000 65,000 

Elongation in 2" (per cent) 9 27 

Reduction in Area (per cent) 24 59 

Brinell Hardness 490 175 

(Values between these extremes may be attained by variations in heat treatment) 

(2) An alloy that might be called "mild stainless steel." This has 
the same chromium content as the preceding steel, but because of 



April, 1935] SYMPOSIUM ON MATERIAL FOR EQUIPMENT 349 

definitely lower carbon content (usually 0.10 per cent or less) is a 
softer and more easily workable steel. It is suitable for screws, bolts, 
nuts, shafting, or other mechanical parts not subject to the most se- 
vere service. This steel is produced in two modifications: regular 
and "free-machining." In the latter, other elements (usually zir- 
conium sulphide) are added in small proportion to render the steel 
free-cutting so that it may be used in automatic screw machines for 
rapid production of small parts in quantity, such as screws, nuts, 
studs, etc. 

This steel is also produced in flat rolled products as sheets, plates, 
and strip, but its corrosion resistance to chemical agents is rather 
moderate, so it is not recommended for applications that demand 
highest corrosion resistance. Table II gives average physical 
properties. 

TABLE II 

f Properties of "Mild Stainless Steel" 

'. Heat Treated Annealed 

Ultimate Strength (Ib./sq. in.) 198,000 85,000 

Yield Strength (Ib./sq. in.) 179,000 55,000 

Elongation in 2" (per cent) 19 30 

Reduction of Area (per cent) 50 72 

Brinell Hardness 395 170 

(3) The next steel to consider contains a higher proportion of chrom- 
ium than either of the preceding, namely 16-18 per cent, and carbon is 
maintained less than 0.10 per cent. The low carbon with the higher 
chromium results in a softer and more ductile alloy. This steel 
might be called stainless "iron" in that it has virtually no hardening 
capacity and in that respect more nearly resembles an iron in its 
softness and ductility. It is to be understood, however, that it is 
by no means a weak material. Its comparative softness is responsible 
for its use in sheet or strip form for the manufacture of articles and 
parts that are formed by bending, pressing, and deep drawing. It 

TABLE m 
Properties of Annealed, High Chromium, Low- Carbon Steel 

Ultimate Strength (Ibs./sq. in.) 80,000 

Yield Strength 55,000 
Elongation in 2" (per cent) 30 

Reduction of Area (per cent) 55 

Brinell Hardness 165 



350 W. M. MITCHELL [J. S. M. P. E. 

possesses high corrosion resistance, and the properly polished surface 
presents a beautiful silvery luster that is untarnishable when exposed 
to the atmosphere. Table III gives average physical properties. 

(4) The last steel to be considered for structural applications is 
"18-8." Although this steel should more properly be classed in the 
corrosion resisting group, it may equally well be used for machine 
parts that require high resistance to corrosive attack. It is produc- 
ible in all desired forms sheets, bars, plates, strip, seamless and 
welded tubing, wire, etc. For various applications there are small 
variations in analysis, and we may recognize three types : regular 
used chiefly for corrosion resistance; free-machining a special 
analysis containing a small percentage of selenium to impart easier 
machining qualities, as this steel will sometimes prove difficult to 
handle in machining, drilling, etc.; and a "stabilized" analysis con- 
taining a small percentage of titanium, and used practically entirely 
as a corrosion -resist ing metal for tank work, which must be fabricated 
by welding. 

This steel has high ductility and toughness as shown by its physical 
properties, which average as in Table IV. 

TABLE IV 

Properties of "18-8" Steel 

Ultimate Strength (Ib./sq. in.) 85,000 

Yield Strength (Ib./sq. in.) 40,000 
Elongation in 2" (per cent) 60 

Reduction of Area (per cent) 70 

Brinell Hardness 135 

Freedom from tarnishing in the atmosphere is one of the important 
characteristics of the stainless steels. This holds true in the salt air 
of the seashore and the moist air of the tropics, as well as in less severe 
surroundings. It should be emphasized that this property is inherent 
in the metal itself, and that it is tarnish resistant "all the way 
through." It is thus not a mere surface coating, as produced by 
electroplating operations, which may peel or be worn off in service. 
This is a fundamental property, and one that gives these steels an 
outstanding position in this respect. Another important fact to be 
noted is that all varieties of stainless steel are superior to regular 
carbon steel in resisting heat, both from the standpoint of resistance 
to scaling and maintenance of strength; hence, they are particularly 
applicable to the construction of projection apparatus. 



April, 1935] SYMPOSIUM ON MATERIAL FOR EQUIPMENT 351 

In considering the second group of applications, namely, for corro- 
sion resistance, we shall disregard chemical plant operations which 
are concerned with the manufacture of motion picture film, silver 
halide emulsions, etc., as this is a phase of the industry which is more 
chemical than photographic. 

Corrosion-resisting applications in the motion picture industry are 
mainly concerned with equipment for fixing and development of 
positive and negative film. Such equipment consists of tanks, shaft- 
ing, and various intricate devices that convey the film through the 
processing solutions and drying chambers. There are several alloys 
suitable for the manufacture of this equipment, viz., "18-8" and 
"18-8" with molybdenum. 

(1) "18-8." The 18 per cent chromium, 8 per cent nickel steel 
already mentioned. This is produced by steel mills in sheets, plates, 
strip, seamless and welded tubing, wire, bars, etc. As this steel can 
can be formed by pressing, bending, and deep drawing, and fabricated 
by all methods, it is possible to construct from it virtually any kind 
of equipment or apparatus that is desired. It has higher corrosion 
resistance than any of the straight chromium steels previously men- 
tioned. General resistance is against nitric acid and oxidizing agents, 
sulphur and many of its compounds, and to many organic acids and 
their salts. The publication of data derived from laboratory corro- 
sion tests is considered inadvisable, as such tests rarely reproduce 
actual working conditions and hence are limited in ability. A better 
method of determining the value of any alloy for corrosion resisting 
purposes is by test under actual working conditions, and such tests 
can usually be made economically and with reasonable convenience. 

Actual experience with the usual developing baths containing pyro, 
metol (elon), and hydroquinone, alone or in combination, shows that 
"18-8" is unattacked by these solutions. In consequence, this steel 
is now practically standard throughout the larger studios for con- 
struction of developing machines and parts. 

For bleaching solutions containing potassium ferricyanide and for 
toning solutions, "18-8" has not been altogether satisfactory when 
the solutions are allowed to stand in trays exposed to the air for in- 
definite periods. The same applies to fixing baths under certain con- 
ditions, where attack may take place at the air-liquid surface. For 
such chemical media a still more corrosion resisting steel may be re- 
quired, such as "18-8 Mo." 

(2) "18-8 Mo." This steel was also developed by Strauss of the 



352 W. M. MITCHELL [j. . M. p. E. 

Krupp Gun Works, and was originally applied to the sulfite pulp in- 
dustry, where corrosion is particularly severe. The addition of some 
2-4 per cent molybdenum to the "18-8" analysis confers additional 
resistance to a variety of chemical media, particularly acids, both or- 
ganic and inorganic. In consequence, this steel is rapidly coming 
into use in many industries, and so far it has given excellent results. 
"18-8 Mo" is produced in flat rolled products, as sheets, plates, and 
strip, and also in bar stock and in castings. The production of seam- 
less tubing is, however, difficult and has not yet reached the com- 
mercial stage. 

Another steel, also containing molybdenum, which may be con- 
sidered a modification of "18-8 Mo," has the composition 27 per cent 
chromium, 3.5 per cent nickel, and 1.5 per cent molybdenum. It has 
the advantage of easier workability and has successfully been pro- 
duced in seamless tubing. It is quite probable that either or both 
these stainless steels with molybdenum will be found resistant to 
bleaching and toning solutions and to fixing baths. Either can be 
welded satisfactorily, and as molybdenum acts as a "stabilizing" 
agent, heat-treatment of welded equipment after fabrication will 
probably not be required. 

(3) The U. S. Steel Corporation has recently placed a new product 
on the market which is particularly well suited for construction of de- 
veloping tanks. This is a stainless veneered steel, to which the 
trade-name Plykrome has been given. It is essentially a composite 
metal made up of a plate of carbon structural steel to which has been 
bonded a relatively thin coating of stainless steel of the "18-8" analy- 
sis. These are perfectly united by a patented process, and will remain 
so under all ordinary fabricating operations. Compared to solid 
stainless steel plate, which is corrosion resistant on both sides, Ply- 
krome is corrosion resisting on one side only. As it may be welded, 
with proper technic, so that the weld is as corrosion resistant as the 
rest of the surface, it may be fabricated into tanks of all kinds. Its 
base price is about one-half that of solid "18-8" plates of the same 
thickness. Thus, it is possible to effect a saving of from 25 to 40 per 
cent more, depending upon labor cost, over the cost of solid stainless 
tanks of identical dimensions, by the use of Plykrome. 

DISCUSSION 

CHAIRMAN CRABTREE: How do the corrosion resisting properties of free 
machining steel compare with those of 18-8? How is it made free-machining? 



April, 1935] SYMPOSIUM ON MATERIAL FOR EQUIPMENT 353 

MR. MITCHELL: By adding a small amount of selenium. Selenium is analo- 
gous to sulphur, chemically. It makes the steel, I won't say easy to cut, but 
easier to cut then the ordinary 18-8. 

CHAIRMAN CRABTREE: What are its corrosion resisting properties? 

MR. MITCHELL: Virtually the same. The difficulty of machining 18-8, as 
those of you know who have had any experience with it, is due to the work-harden- 
ing properties of the metal. The metal hardens rather severely when worked 
cold, and if the tool is at all dull, the surface of the metal will become hardened and 
glazed so it becomes impossible to cut through it. 

To avoid that, the tool must be sharp, have ample clearance and "lip-rick," 
as we say, and plenty of power behind it. The free-machining variety, of course, 
is of more use for work in automatic screw machines where large numbers of small 
parts are to be made. 

CHAIRMAN CRABTREE: I understand that when you weld the 18-8, unless 
particular precautions are taken to anneal the weld, the weld will corrode more 
rapidly than the material proper. Do I understand that there is a new variety 
which is more satisfactory for welding and which is not open to this objection? 

MR. MITCHELL: Yes. The difficulty is not corrosion at the weld itself but in 
the region adjacent to the weld. It can be avoided to a certain extent by specify- 
ing a material of low carbon content. The difficulty is caused by the precipita- 
tion of chromium carbide which takes place at certain temperatures in regions 
adjacent to the weld. The formation of chromium carbide absorbs chromium 
from the solid solution of the steel, so lowering its corrosion resistance. This oc- 
curs particularly along the borders of the crystalline grains, so that the material 
will be disintegrated completely under certain conditions. That can be over- 
come by using a very low carbon material, or by heat-treating the entire piece, 
which is very often impracticable because of the general size or shape. This 
difficulty has also been overcome by adding a small quantity of titanium or 
columbium. Carbon has greater affinity for titanium or columbium than for 
chromium, and under favorable conditions there will be precipitation of titanium 
or columbium carbide instead of chromium, and the corrosion resistance of the 
metal is not destroyed thereby. 

This material is called "stabilized" because it is given a special heat-treatment 
which stabilizes it against change during welding. It has been found to be quite 
successful and obviates the former difficulty of corrosion alongside the welds in the 
regular 18-8. 

CHAIRMAN CRABTREE: With the low carbon content material, what do you 
sacrifice by lowering the carbon? In other words, why not always use low 
carbon? 

MR. MITCHELL: The sacrifice is on the part of the steelmaker, because the low 
carbon is more difficult and more expensive to produce. 

CHAIRMAN CRABTREE : Can you say a little more about the properties of the 
18-8 with molybdenum? You say it is more resistant to corrosion. Is it more 
difficult to machine? 

MR. MITCHELL: I should not say it is any more difficult to machine. Molyb- 
denum was introduced originally for the sulfite pulp industry, to provide better 
resistance against the sulfurous acids and other conditions encountered in the 
digester and pipe lines. There it has given very good service. Due to the fact 



354 R. L. FOOTE [J. S. M. P. E. 

that it has better resistance to dilute mineral acids than the straight 18-8, it has 
found various applications in chemical plants. Also it has been used successfully 
in the textile industry, where dilute acid solutions are used in connection with 
dyeing. 

I don't know whether it has been used at all in photographic work, but it would 
seem to me that due to its higher corrosion resistance than the straight 18-8, it 
should be a good material to consider for fixing-tanks and so forth, where there has 
been trouble with 18-8. 

CHAIRMAN CRABTREE : What happens when you weld it ? 

MR. MITCHELL: It can be welded as satisfactorily as 18-8, and there is very 
little difficulty from carbide precipitation ; it must be used also in the low carbon 
form. It has excellent physical properties, is amply strong for any requirements, 
and can be obtained in the form of sheets or bars, etc. Seamless tubing is rather 
difficult to produce, and is not generally available. The price of 18-8 with added 
molybdenum is about one-third more than the regular 18-8. 



LAMINATED BAKELITE IN THE MOTION PICTURE INDUSTRY 

R. L. FOOTE* 

Summary. The nature, manufacture, and applications of laminated bakelite are 
described briefly, and reference is made to its current and possible uses in motion 
picture work. 

Laminated bakelite has a definite set of unusual physical and chemi- 
cal properties quite unlike those of any other material. It will be 
shown later, how certain of those properties combine to make lami- 
nated bakelite a material of unusual versatility and value in mechanical 
applications. 

In the manufacture of laminated bakelite a base material such as 
kraft paper, cotton rag paper, cotton duck of various thicknesses, 
asbestos paper or asbestos fabric, is first impregnated with a dis- 
solved or liquid phenol formaldehyde synthetic resin. This is a 
continuous process in which the base material is handled in rolls and 
is continuously impregnated and dried. 

During the coating operation, it is necessary to maintain careful 
laboratory control of the specific gravity of the resin and finish with 
an exact degree of dryness. If the equipment that is used is modern 
and accurate, and the operating personnel careful and conscientious, 
the resulting product will be a strong, uniform, stable, moisture- 
resistant and free-machining laminated bakelite. Inferiorities in 

* Synthane Corp., Oaks, Pa. 



April, 1935] SYMPOSIUM ON MATERIAL FOR EQUIPMENT 355 

either equipment or men will result in material that may readily 
absorb moisture, machine poorly, warp badly, have low shock- 
resistance or perhaps most serious of all be subject to change of 
state; that is, the material will show definite changes in its properties 
after shipment and when subjected to normal service conditions. 

For the production of laminated bakelite sheet, the coated sheet 
material is next cut to size. The desired number of coated sheets 
are stacked between polished stainless steel finishing plates and 
placed in a hydraulic press having steam-heated platens. The 
stack or make-up, as it is called, is then cured at a temperature of 
about 325F. under a pressure of 2000 pounds per square inch. 

Though nearly dry, the synthetic resin in the coated sheets will 
soften under the influence of heat. It is however, thermo-setting; 
that is, the continued application of heat causes a chemical change to 
occur, when the resin becomes a hard, infusible, insoluble substance 
that is definitely not thermoplastic. The heavy pressure causes the 
"make-up" to compress as the resin softens, and to have a minimal 
cross-section when the chemical change, or "kick-over," as it is 
termed, occurs. This results in the product known as laminated 
bakelite sheet. Similarly > curing the coated stock after first winding 
it upon a mandrel of the desired size and shape, produces tubular 
laminated bakelite. The resulting product may be sawed, turned, 
drilled, punched, milled, shaped, or threaded. Any of these machining 
operations may be performed with comparative ease, and should be 
done at relatively high speed. 

In the motion picture industry many uses of laminated bakelite are 
already common practice. Throughout the sound equipment field, 
laminated bakelite is widely used for insulation, in both recording and 
reproducing apparatus. In addition, gears, pinions, shims, and bush- 
ings for cameras and projectors are at times made of this material. 

With quietness becoming an increasingly important factor in such 
equipment, still better silent gear materials are required. Some 
refinements have already been made, but still greater improvement is 
possible. Materials quite in advance of anything commercially 
available are now possible in the laboratory and will soon be offered 
for these purposes. 

In the processing equipment field, the usual electrical applications 
to panels, insulating washers, and so forth, are to be found, of course. 
Much less common, but of great importance, are the purely mechani- 
cal applications recently developed. Laminated bakelite combines 



356 R. L. FOOTE [J. S. M. P. E 

high ratio of strength to weight, ready machinability, and quite com- 
plete resistance to the action of all chemicals thus far encountered in 
film processing. Thus, it is an ideal material to use for stationary 
parts in contact with otherwise corrosive chemicals. These parts 
have so far included various bushings, spacers, frame members, pipe 
fittings, and other stationary parts. In addition, laminated bakelite 
makes an ideal bearing or journal when lubricated with water or what- 
ever chemical solution might be present. Free-running and long- 
wearing bearings and idler rolls are being applied more and more 
widely. 

'Some doubts have existed as to the material's ability to resist the 
action of caustic soda. Due, however, to the low concentrations 
generally employed, the development of caustic resistant resins, and 
improved technic in manufacture and application, this difficulty has 
been completely overcome. 

Laminated bakelite possesses several other interesting features. 
As compared with metals, it is nearly non-resonant. To improve its 
sound and vibration absorbing qualities still further, a method of 
laminating sheet rubber into the make-up has been developed. 
This results in an almost completely non-resonant material, yet one 
that retains the physical strength of laminated bakelite. It is 
possible also to treat the surface sheet so as to imitate wood grains, 
leather surfaces, pebble finish, or even weather-beaten shingles. 

It is probably impossible permanently to remove all noise from a 
rotating mechanism, such as a gear train. Refinement of the mecha- 
nism soon reaches a point of commercial impracticability. But 
if we wish completely to silence a camera or projector mechanism of 
this sort, why not house it in a non-resonant case of this material with 
a surface finish of imitation morocco leather or pebble grain? 

Laminated material can be molded in irregular cross-sections, and 
yet retain high impact strength. In order to stiffen a simple beam 01 
cantilever section, it is possible to bury a metal insert in the section 
Again, with some limitations, we may combine the strength and 
stiffness of metal with the corrosion-resisting properties, permanent 
finish, and, perhaps, eye-appeal of laminated bakelite. 

DISCUSSION 

MR. MACOMBER: How does bakelite withstand caustic soda, sulfuric acid, 
and acetic acid, as compared with rubber? 

MR. FOOTE: Sulfuric acid, up to about 20 per cent concentration, shows no 
effect, and is not materially detrimental. I know of no case where acetic acid 



April, 1935] SYMPOSIUM ON MATERIAL FOR EQUIPMENT 357 

has shown any effect upon a good piece of laminated stock. As to caustic soda, 
if we are not interested in the surface appearance, and the concentrations used 
are not too high, it is quite satisfactory. For example, we have some heavy-duty 
bearings working in about 14 per cent caustic and at very high unit loads and 
low speeds. After a year or so of service, they look perfectly awful when they 
are taken out, but are entirely serviceable as bearings. 

It is now possible to produce material that is approximately ten times as 
caustic-resistant as, let us say, the ordinary garden variety of laminated bakelite 
was a year or so ago. Whereas we used to make a piece that would last two years, 
we can now make it last ten times as long, according to tests. 

CHAIRMAN CRABTREE: With regard to ordinary photographic solutions, 
that is, developers and fixing solutions, these materials are surprisingly resistant. 
In cases where slight possible distortion or expansion of the material is not im- 
portant, I believe the material is a very promising one for such equipment. 



INCONEL AS A MATERIAL FOR PHOTOGRAPHIC FILM PROCESSING 

APPARATUS 

F. L. LAQUE* 

Summary. The chemical and physical properties of the alloy Inconel are de- 
scribed, with especial reference to the requirements of film processing equipment. 
Examples of applications of Inconel in motion picture film handling apparatus are 
given. Asa background to the use of Inconel, the more familiar properties and uses of 
monel metal and pure nickel with photographic solutions are discussed briefly. 

Inconel is an alloy first made commercially available by the 
International Nickel Company, Inc., in 1931. Since it is one of the 
newest of the corrosion-resisting materials, a brief description of its 
general characteristics will be given before a discussion of its useful- 
ness in connection with film processing. 

COMPOSITION AND MECHANICAL PROPERTIES 

Inconel contains approximately 80 per cent of nickel, 14 per cent of 
chromium, and 6 per cent of iron. Its principal constituents combine 
the acid and alkali resistance of nickel with the passivating effect of 
chromium to provide resistance to the attack of a wide variety of 
corrosives, including those encountered in film treatment processes. 
It is pertinent to the present discussion that nickel, the principal con- 

* Development & Research Department, International Nickel Company, Inc., 
New York, N. Y. 



358 



F. L. LAQuE 



[J. S. M. P. E. 



stituent of Inconel, has for many years been used to protect films and 
emulsions from harmful metallic contamination during their manu- 
facture. Typical of such use of nickel is the emulsion strainer illus- 
trated in Fig. 1. 

The mechanical and physical properties of Inconel are given in 
Tables I and II, from which it will be seen that the alloy is excep- 
tionally strong and ductile, and that the range of both strength and 
ductility provide an opportunity for selection of a combination 




FIG. 1. Pure nickel photographic emulsion strainer. 

best suited to particular needs. Thus, for example, the ductility 
desirable for sheet fabrication is as readily attained as the high 
strength required for springs. The strength and hardness properties 
of Inconel give it good resistance to wear, which is often needed for 

TABLE I 

Mechanical Properties of Inconel 

Tensile Strength Yield Point Elong. 



Sheet and strip 

Annealed 
Rod 

Annealed 

Cold Drawn 
Wire 

Annealed 

Spring Temper 



(lbs./in.*) 

80-95,000 



(lbs./in.*) 

30-40,000 



Red. in Area 
(per cent) (per cent) 



80- 95,000 30-40,000 

100-115,000 80-95,000 

80- 95,000 30-40,000 
175-200,000 



45-55 

45-55 
20-30 

45-55 



65-75 



April. 1935] SYMPOSIUM ON MATERIAL FOR EQUIPMENT 



359 



TABLE II 



Physical Constants of Inconel 



Density 

Coefficient of expansion 
100-200F. range 

per F. 

per C. 
100-1400F. range 

per F. 

per C. 

Heat conductivity 
Specific heat (77-212F.) 
Melting point 



Modulus of elasticity 
Modulus in torsion 



8.55 



0.0000064 
0.0000115 

0.00000696 
0.0000161 

3.5% that of copper 
0.109 

2,530 

1,388 

32,000,000 
10,300,000 



parts rubbed by rapidly moving film. The high value of the modulus 
of elasticity indicates exceptional rigidity; and the low coefficient 
of expansion, which, incidentally, is about the same as that of steel, 
is useful in welding operations (by reducing buckling) and in operating 
equipment subjected to wide or frequent changes of temperature. 

The widespread use of springs in cameras, projectors, and film 
developing equipment makes the excellent spring properties of 
Inconel especially interesting. These properties are given in Table 
III, which includes also, for convenient reference, data on some other 

TABLE III 
Spring Properties of Inconel 



Metal 



Ferrous 

Music wire 
Carbon steel 
Chrome vanadium 
Tungsten steel 
(high alloy) 
Stainless Steel 

Nonferrous 

Inconel 
Monel Metal 
Phosphor bronze 
Brass 



5 

Torsional 
Elastic 
Limit 

(lbs./in.*) 



G 

Torsional 
Modulus 



S-G S*G 

Deflection Stored 
Index Energy 
Index 



Maximum 

Operating 

Temperature 



140,000 12,000,000 

125,000 11,500,000 

110,000 11,500,000 

110,000 11,500,000 

100,000 10,000,000 



100,000 10,300,000 0.0097 960 

70,000 9,000,000 0.0072 540 

70,000 6,500,000 0.011 750 

50,000 5,500,000 0.0091 450 



0.012 
0.011 
0.0096 


1640 
1360 
1050 


200 
400-430 
400-430 


0.0096 
0.010 


1050 
1000 


700 
500-600 



600-700 

400 

Atmospheric 
Atmospheric 



360 F. L. LAQUE [J. S. M. P. E 

commonly used spring materials. The torsional elastic limit of 
Inconel given in Table III refers to coiled helical springs after solid 
compression; as coiled, the springs have a torsional elastic limit of 
65,000-75,000 pounds per square inch. 

Inconel is available in all the usual forms, such as sheets, bars, 
rods, tubes (seamless and welded), and wire. It is amenable to 
fabrication by all common methods and can be welded, soldered, and 
silver-brazed. Its corrosion resistance is not adversely affected by the 
heat of welding, and welded parts do not require heat treatment to 
assure corrosion -resistance. 

GENERAL CORROSION-RESISTANCE 

The corrosion-resistance of Inconel is, of course, of greatest interest 
to motion picture engineers. The alloy was developed primarily to 
resist corrosion and staining by organic acids such as are found in 
fruits and other food products. To illustrate its resistance to dilute 
acetic acid, commonly used in film processing solutions, specimens of 
Inconel were exposed to vinegar containing 10 per cent of acetic acid 
in a storage tank for 240 days. The indicated rate of attack upon 
specimens subjected to both continuous and intermittent immersion 
was equivalent to penetration at a rate of less than one ten-thousandth 
of an inch per year. In boiling 2 per cent acetic acid the rate of corro- 
sion of Inconel was found to be less than one-thousandth of an inch per 
year. Data on the resistance of Inconel to sulphuric, hydrochloric, 
and nitric acids are given in Table IV. 

TABLE IV 

Rates of Corrosion of Inconel by Sulfuric, Hydrochloric, and Nitric Acids 



Acid Rate of Corrosion in 


Conditions of Test 


Inches Penetration 




per Year* 




5% Hydrochloric 0.125 


Cold, fully aerated, 16 ft. per minute flow 


5% Sulfuric 0.125 


Cold, fully aerated, 16 ft. per minute flow 


5% Sulfuric 0.009 


Cold, air free, 16 ft. per minute flow 


5% Sulfuric 0.200 


Hot, fully aerated, 16 ft. per minute flow 


5% Nitric 0.22 


Cold, fully aerated, 16 ft. per minute flow 


25% Nitric 0.002 


Cold, fully aerated, 16 ft. per minute flow 


45% Nitric 0.001 


Cold, fully aerated, 16 ft. per minute flow 


05% Nitric 0.003 


Cold, fully aerated, 16 ft. per minute flow 



* The depth to which uniform corrosion would penetrate a surface of the 
metal exposed to the corroding solution continuously for a year. 



April, 1935] SYMPOSIUM ON MATERIAL FOR EQUIPMENT 361 

Inconel possesses sufficiently good resistance to sulfuric and 
hydrochloric acids to enable its use where these acids may be en- 
countered. However, the availability of such other materials as 
monel metal or nickel, whose resistance to these acids is superior to 
that of Inconel, would not justify its use, simply because of its resis- 
tance to sulfuric or hydrochloric acids. At the same time, the fact 
that Inconel withstands such a highly corrosive solution as aerated 
5 per cent hydrochloric acid as well as it does, is impressive evidence of 
its inherent resistance to chemical attack. 

It will be noted also in Table IV that the resistance of Inconel to 
concentrated nitric acid is superior to its resistance to the dilute 
(5%) acid. Consequently, if nitric acid should be used for cleaning 
Inconel film processing equipment, it would be preferable to use acid 
of 25 per cent strength or stronger. 

Because of its high nickel content Inconel is, of course, strongly 
resistant to attack by alkaline solutions, while its chromium content 
increases its resistance to sulfur compounds. Consequently, it 
would be expected that Inconel would be highly resistant to alkaline 
sulfur compounds. That this is the case was demonstrated by a 
test in a boiling concentrated (50%) solution of sodium sulfide, where 
the indicated rate of attack was equivalent to penetration at a rate 
of only about four one-thousandths of an inch per year. The re- 
sistance of Inconel to straight caustic corrosion was indicated by a 
test in which specimens were exposed for thirty days within a vacuum 
evaporator producing caustic soda of 70 per cent strength at 260 F. 
The results showed that Inconel was practically free from attack, the 
rate of penetration being less than one ten-thousandth of an inch per 
year. 

RESISTANCE TO DEVELOPING SOLUTIONS 

It is not only necessary that a material used in contact with photo- 
graphic solutions be resistant to chemical attack by them, but it is 
also requisite that the material be free from detrimental effects upon 
the properties of the solutions themselves. The advantage of nickel 
and high nickel alloys in this respect has already been mentioned in 
connection with the manufacture of photographic film. 

Of probably greatest importance from the standpoint of metallic 
contamination of film processing solutions is the well-known effect of 
certain elements in inducing fogging action in developers. This 
subject has been investigated in considerable detail by Crabtree and 
Matthews 1 who found that copper and tin and alloys containing these 



362 F. L. LAQuE [J. s. M. P. E. 

elements were most active in inducing chemical fog in developing solu- 
tions. A notable exception to this general conclusion was that monel 
metal, although containing copper, did not fog the developing solu- 
tion. These investigators found also, by adding metal salts to a 
typical developer, that tin and copper had a fogging action, but that, 
among other elements tested, nickel and iron constituents of In- 
conel had no fogging effect. Similar data on chromium, the third 
constituent of Inconel, are lacking, but sufficient experience has been 
obtained to show that chromium, as it is present in Inconel, at least, 
has no detrimental effect upon developing solutions. 

Quantitative data on the resistance of Inconel to developing solu- 
tions are not available, but results of qualitative tests and practical 
use of the material in the form of containers for such solutions have 
demonstrated that corrosion is practically nil. This is not surprising, 
in view of the widespread use of monel metal and nickel for handling 
developing solutions, and the fact that the resistance of Inconel to the 
chemicals used in developers is at least as good as that of either monel 
metal or nickel. 

Typical of results of practical experience with Inconel in contact 
with developing solutions is the following quotation from a report 
from the first user of Inconel for photographic solutions: "For the 
past six months we have been using trays made of chrome nickel 
(Inconel) for developing prints, and find that they are very satisfac- 
tory indeed. The developer forms a deposit upon the inside surface 
of the tray, but this deposit is very easily removed with 28 per cent 
acetic acid, and there is no corrosion whatever of the surface of the 
tray." On several occasions Inconel has been tested as to its suit- 
ability for use with developing solutions, with uniformly satisfactory 
results, as indicated by the following examples of such tests : 

A specimen of Inconel was exposed for more than six weeks in a 
standard solution of Eastman-prepared x-ray developer at the Uni- 
versity Hospital, University of Michigan. This sample showed not 
the slightest evidence of corrosion after this test. Specimens of 
Inconel were exposed to developing solutions (elon-hydroquinone 
and pyro-soda) at Wright Field, Dayton, Ohio, for more than ten 
months, without any evidence of corrosion. Tests in the laboratories 
of a large producer of photographic equipment demonstrated that 
Inconel is satisfactory for use with developing solutions. Of direct 
interest to motion picture engineers is that at least four of the largest 
motion picture companies on the Pacific Coast have made exhaustive 



April, 1935] SYMPOSIUM ON MATERIAL FOR EQUIPMENT 363 

tests on the suitability of Inconel for use with developing solutions, 
and all have found it to be entirely satisfactory. 

FIXING SOLUTIONS 

Probably the most important property of Inconel, so far as equip- 
ment for film processing is concerned, is its resistance to the complex 
corrosive effects of wholly, or partly, exhausted hypo fixing solutions. 
It is a well-known fact that most metals and alloys will precipitate 
silver from such solutions. Such precipitation of silver is always ac- 
companied by a certain amount of corrosion, part of which is due to 
the direct solution of an equivalent amount of the material to replace 
the silver plated out, and part to complex galvanic and concentration 
cell effects that follow the deposition of silver and the entrapment 
of fixing solution beneath loose deposits. Crabtree, Hartt, and 
Matthews 2 have shown that such silver deposits are not protective to 
the underlying metal. 

It has been found that under ordinary conditions of use no silver 
will deposit upon Inconel in fixing solutions, and thus the corrosion 
is practically nil. This property is also important in connection with 
moving parts, the dimensions of which must be maintained within 
very close limits and where the deposition of silver may cause me- 
chanical difficulties as well as corrosion. 

There is some indication from tests in the laboratories of a film 
producer that the deposition of silver on Inconel which does not occur 
in fixing solutions at ordinary temperatures may occur at higher 
temperatures. In some tests in a used fixing bath at 110F., a slight 
amount of silver was plated out upon an Inconel tray used to hold the 
solution, but there was no appreciable corrosion of the tray. 

It is interesting to note that the first piece of Inconel used for 
photographic equipment was made up into a welded container for a 
hypo solution. This container has been in practically continuous use 
for about three years, and has remained bright and free from silver 
deposits and shows no evidence of corrosion. A specimen of Inconel 
was exposed for more than six weeks in a standard hypo fixing solution 
at the University Hospital, University of Michigan. It remained 
clean and bright, and free from deposited silver. A sample of 
Inconel that had been continuously immersed in hypo for ten months, 
in the film developing department of the International Nickel Com- 
pany's Research Laboratory at Bayonne, N. J., also showed no signs 
of corrosion or deposited silver. 



364 F. L. LAQUE [j. s. M. P. E. 

Most of the larger motion picture studios have tested Inconel in 
contact with hypo fixing solutions, and all have reported that it has 
withstood their tests satisfactorily. A typical result of such a test 
follows : 

A specimen of Inconel in the form of a sheet measuring 2 1 /* by 3 l / 2 
in. was suspended in the reservoir tank of the hypo circulation system 
in the film developing laboratories of Consolidated Film Industries, 
Fort Lee, N. J. It was so located that part of the time it was im- 
mersed in the solution and part of the time exposed to the air. The 
sample was weighed before exposure, after eight days in the tank, and 




FIG. 2. Monel metal shafting, gears, bolts, and screws in main 
channel of film-developing unit. 



at the end of the test period of one hundred and fifty-four days. 
The loss of weight during the first eight days was only 0.7 mg., and 
the total loss of weight during the whole test was only 2.5 mg., or 
less than 0.0001 oz. That such a rate of corrosion is insignificant 
can be readily appreciated when it is considered that it would ac- 
count for penetration of a sheet to a depth of only 0.000003 inch per 
year. 

Its better behavior in fixing solutions is the principal reason why 
Inconel should be preferred to the older materials, monel metal and 
pure nickel, for film processing equipment. In fact, so far as de- 
veloping solutions are concerned, both nickel and monel metal are 
entirely adequate, and it is doubtful whether Inconel or any other 
material could show an important advantage over them. The older 



April, 1935] SYMPOSIUM ON MATERIAL FOR EQUIPMENT 365 

materials are also suitable for fixing solutions in amateur work where 
the silver content of the fixing baths does not build up to as great an 
extent as in professional or commercial laboratories. Both monel 
metal and nickel are useful for wash-tanks. However, for hypo wash- 
tanks Inconel would have the advantage of remaining bright, whereas 
nickel and monel metal would gradually become darkened by silver 
and sulfur compounds. 

Monel metal has been used successfully to a considerable extent 
for motion picture developing apparatus in the form of shafting, 
gears, rods, bolts, etc., on automatic equipment such as is illustrated in 
Fig. 2. Other uses for both monel metal and nickel are trays, clips, 
etc., such as illustrated in Fig. 3. 

TONING, INTENSIFICATION, AND REDUCTION SOLUTIONS 

Toning baths, as a class, appear to be much more corrosive to 
metals than either developing or fixing solutions. This is especially 
true of solutions containing potassium ferricyanide and salts of iron 
and uranium. Consequently, caution should be exercised in using 
them in equipment made of Inconel or other corrosion-resisting mate- 
rials. At the same time, it is interesting to note that Inconel has 
been found to be much superior to monel metal for this service. An 
Inconel coil has been in use for several months for heating and cool- 
ing a sepia toning bath containing hypo, potassium alum, silver 
nitrate, and silver chloride in the usual proportions. It is true that 
the Inconel has suffered some corrosion, as evidenced by the ac- 
cumulation of a scale of corrosion product plus some deposit from the 
solution, most evident at the liquid line. However, the rate of attack 
has been moderate, and the accumulated scale appears to have been 
somewhat protective. Glass coils which the Inconel replaced were 
less satisfactory because of occasional breakage and the resulting 
detrimental effects upon the prints being processed. It is significant 
that whatever corrosion of the Inconel occurred had no harmful 
effect upon the quality of the work turned out. 

Solutions used for intensification, such as those containing bichlo- 
ride of mercury, are also very corrosive to metals. Inconel is not 
recommended for use in contact with solutions containing appreciable 
percentages of bichloride of mercury. 

Inconel should be satisfactory for use with bleaching solutions such 
as a mixture of potassium bichromate and sulfuric acid at atmos- 
pheric temperature. A sample of Inconel showed no evidence of 



366 



F. L. LAQUE 



[J. S. M. P. E. 



attack after three days' contact with a solution made up to contain 
3 3 /4 oz. of potassium bichromate and 5 oz. of sulfuric acid in a gallon 
of water. 

Solutions used for reduction also belong in the group that require 
caution in their use with metals. This is especially true of those con- 
taining free iodine, which should not be used in contact with Inconel 
or other metals. Data on the behavior of Inconel in contact with 
acid solutions containing potassium permanganate and ammonium 
persulfate are lacking, but in view of its good behavior in the test 
with the acid bichromate mixture mentioned earlier, it is possible 



CC 




J 



FIG. 3. Monel metal film developing clips. 

that it will be found satisfactory for use with these other oxidizing 
acid solutions. However, it should be used with caution with these 
solutions until its suitability has been definitely established. 

Data on bleaching solutions are contradictory, in that tests con- 
ducted in a film company's laboratories have indicated the unsuit- 
ability of Inconel for use with bleaching solutions containing potas- 
sium bromide and potassium ferricyanide, whereas reports from the 
field mention the replacement of less satisfactory materials with 
Inconel for parts of motion picture film processing equipment that 
are used in bleaching solutions. 

GALVANIC BEHAVIOR 

In view of the work done by Crabtree, Hartt, and Matthews 2 on the 
effect of electrolysis on the rate of corrosion of metals in photo- 



April, 1935] SYMPOSIUM ON MATERIAL FOR EQUIPMENT 



367 




FIG. 4. Eight-inch plunger pump lined with Inconel for pumping 

hypo. 




FIG. 5. Inconel spindle and spacers for film rolls in continuous developing 

equipment. 



graphic solutions, it is desirable to establish the position of Inconel in 
an electromotive series referred to developing and fixing solutions. 
Inconel under ordinary conditions of use will occupy a position very 



368 F. L. LAQUE [j. s. M. P. E. 

close to silver in such a series. Since it does not precipitate silver 
from hypo solutions, it must be either very close to or more noble 
than silver in such solutions, although it may be less noble in other 
environments. In common with other high chromium alloys, such 
as the stainless steels, which owe their apparent nobility to what is 
most commonly believed to be an oxide film, destruction of this film 
may cause Inconel to occupy a position in the electromotive series 
near nickel. In setting up such a series, therefore, it is desirable to 
include the high chromium alloys in two locations, the one referring 
to the passive (passive film present), and the other to the active 
(passive film absent) state. It should, of course, be noted that for 
purposes and under conditions for which Inconel and the other high 
chromium alloys are generally used, the active state is only rarely 
encountered. Such a series, based upon many tests and practical 
experiences, is given in Table V. 

TABLE V 

Electromotive or Galvanic Series of Metals and Alloys 
Corroded end (anodic) 

Magnesium 

Aluminum 

Duralumin 

Zinc 
Cadmium 

Iron 

Chromium-Iron Stainless Steel (active) 

Chromium-Nickel-Iron Stainless Steel (active) 

Lead-Tin Solder 

Tin 

Lead 

Nickel 

Inconel (active) 
Brasses 
Bronzes 
Monel Metal 
Copper 

Inconel (passive) 

Chromium-Iron Stainless Steel (passive) 

Chromium-Nickel-Iron Stainless Steel (passive) 

Silver 

Gold 
Platinum 



April, 1935] SYMPOSIUM ON MATERIAL FOR EQUIPMENT 



369 



It will be noted that the materials are arranged in groups. There 
may be a rearrangement of positions within any group, depending 
upon the environment ; but it is unusual for a change of position to 
occur from one group to another, except in the special case of the 
chromium alloys previously discussed. 

A practical application of this series to photographic film develop- 
ing processes is that all materials grouped above silver in the series 
will precipitate silver from ex- 
hausted or partially exhausted 
fixing solutions. The chromium 
alloys grouped with silver ordi- 
narily will not precipitate silver 
except where they may have 
been rendered active by the 
destruction of their passive 
films. As has been shown earlier 
in the discussion of its behavior 
in hypo fixing solutions, Inconel 
tends to remain passive in these 
solutions, and consequently does 
not precipitate silver from 
them. 

As Inconel is below the lead- 
tin solders and the high copper 
alloys in the galvanic series, 
combinations of these alloys 
with Inconel in developing solu- 
tions should be avoided, because 
these other materials may suffer 
accelerated corrosion, especially 
where the area of Inconel ex- 
posed is great as compared with 
that of the other materials. 
This is in accordance with the 

recommendation of Crabtree, Hartt, and Matthews 2 resulting from 
their investigations of the effects of dissolved copper and tin. Tin 
was found to cause fogging, 1 and copper to accelerate the rate of 
oxidation of a sodium sulfite solution 3 and thereby to shorten the 
life of a developer. 




FIG. 6. Submerged type of pump with 
2-inch Inconel pipe connections. 



370 



F. L. LAQuE 



[J. S. M. P. E. 



PRACTICAL APPLICATIONS OF INCONEL 



The many useful properties of Inconel that have been described 
soon led to its adoption for the fabrication of film processing equip- 
ment, especially in the motion picture industry. Typical of such 
uses of Inconel are the following: 

The plunger pump illustrated in Fig. 4 was lined with Inconel 
about a year ago. It is being used for pumping hypo fixing solutions 
at the West Coast plant of Consolidated Film Industries, Inc. 

Paramount Productions, Inc., Los Angeles, Calif., is using Inconel 
with good success for such equipment as spindles and spacers for film 




FIG. 7. Hunter-Pierce film development machine equipped with Inconel. 

rolls in continuous film developing apparatus as illustrated in Fig. 5. 
The submerged pump shown in Fig. 6 has been equipped with 2-inch 
Inconel pipe for handling a hypo solution. The developing machine 
shown in Fig. 7 equipped with Inconel is in use at the Universal 
Pictures Company plant on the West Coast. One of the most interest- 
ing applications of Inconel is as a coil for controlling the temperatures 
of a hypo fixing solution in the plant of Cinecolor Pictures. This 
coil was put into service about ten months ago. 

In addition to such applications as these in the motion picture 
industry, a considerable amount of Inconel has been used for film- 



April, 1935] SYMPOSIUM ON MATERIAL FOR EQUIPMENT 371 

developing equipment in x-ray and commercial laboratories. The 
success of these and other installations of Inconel confirms the 
results of the numerous tests that have demonstrated its resistance to 
photographic solutions. It is expected that its use for such purposes 
will increase as time goes on. 

REFERENCES 

1 CRABTREE, J. I., AND MATTHEWS, G. E.: "The Resistivity of Various Mate- 
rials toward Photographic Solutions, " Ind. & Eng. Chem., 15 (July, 1923), No. 7, 
p. 666. 

2 CRABTREE, J. I., HARTT, H. A., AND MATTHEWS, G. E. : "Effect of Electroly- 
sis on the Rate of Corrosion of Metals in Photographic Solutions," Ind. & Eng. 
Chem., 16 (Jan., 1924), No. 1, p. 13. 

3 CRABTREE, J. I.: "Chemical Fog," Brit. J. Phot., 66 (Feb., 1919), p.97. 

DISCUSSION 

CHAIRMAN CRABTREE: What is the resistivity of welded Inconel? 

MR. LAQUE: So far as we have been able to determine, neither the deposition 
of the weld nor the heat attending the deposition have any detrimental effect 
upon the corrosion-resistance, either in the weld itself or in the metal adjacent to 
it. We have subjected Inconel to the standard tests used to determine sus- 
ceptibility, and it has come through the tests without any indication that such 
things occur. 

MR. MACOMBER: What is the price of Inconel as compared with that of monel 
metal? 

Mr. LAQUE: In the form of sheets Inconel costs about fifteen cents more a 
pound than monel metal, and about twenty cents a pound in the form of a cubical. 
Castings are always a rule unto themselves. 

CHAIRMAN CRABTREE: Do you find any greater degree of corrosion, at the 
air-liquid boundary? 

MR. LAQUE: If corrosion will occur at all, that will be the place where it 
will begin. Whether it will begin or not in any given case is determined by a 
number of factors that are difficult to control. I should not be surprised that 
under the conditions of use of fixing solutions, where the temperature might be 
higher than normal for appreciable periods, you would find evidence of air-line 
attack. However, I don't believe it would be very serious; it would occur, but 
very slowly. 



SPRING, 1935, CONVENTION 
SOCIETY OF MOTION PICTURE ENGINEERS 
HOTEL ROOSEVELT, HOLLYWOOD, CALIF. 

MAY 20-24, INCL. 
Officers and Committees in Charge 

PROGRAM AND FACILITIES 

W. C. KUNZMANN, Convention Vice-President 
J. I. CRABTREE, Editorial Vice-P resident 
J. O. BAKER, Chairman, Papers Committee 

LOCAL ARRANGEMENTS AND RECEPTION COMMITTEE 

P. MOLE, Chairman 

G. S. MITCHELL E. HUSE G. F. RACKETT 

W. QUINLAN K. F. MORGAN H. W. MOYSE 

J. A. BALL W. C. HARCUS J. A. DUBRAY 

C. W. HANDLEY E. C. RICHARDSON F. E. JAMES 

R. H. McCuLLQUGH R. G. LINDERMAN C. DREHER 

N. LEVINSON 

PROJECTION COMMITTEE 

H. GRIFFIN, Chairman 

J. O. AALBERG R. H. McCuLLoucn 

L. E. CLARK K. F. MORGAN 

Officers and Members of Los Angeles Local 150, 1. A. T. S. E. 

STUDIO AND NEW EQUIPMENT EXHIBIT 

O. F. NEU, Chairman 

H. GRIFFIN J. FRANK, JR. 

P. MOLE S. HARRIS 

BANQUET 

W. C. KUNZMANN, Chairman 

P. MOLE W. QUINLAN E. HUSE 

G. S. MITCHELL G. F. RACKETT O. F. NEU 

PUBLICITY COMMITTEE 

W. WHITMORE, Chairman 

J. J. FINN J. R. CAMERON A. JONES 

F. H. RICHARDSON G. E. MATTHEWS P. A. McGuiRE 

MEMBERSHIP 

O. M. GLUNT, Financial Vice-P resident 
E. R. GEIB, Chairman, Membership Committee 
372 



SPRING, 1935, CONVENTION 373 

LADIES' RECEPTION COMMITTEE 

MRS. E. HUSE, Hostess 

assisted by 

MRS. G. F. RACKETT MRS. F. E. JAMES MRS. E. C. RICHARDSON 

MRS. W. QUINLAN MRS. P. MOLE MRS. F. C. COAXES 

MRS. C. W. HANDLE Y 

Headquarters 

The headquarters of the Convention will be the Hotel Roosevelt, where excellent 
accommodations and Convention facilities are assured. Registration will begin 
at 9 A.M. Monday, May 20. A special suite will be provided for the ladies at- 
tending the convention. Rates for S. M. P. E. delegates, European plan, will be as 
follows : 

Single: $2.50 per day; one person, single bed. 

Double: $3. 50 per day; two persons, double bed. 

Double: $4.50 per day; two persons, twin beds. 

Suites: $6.00 and $8.00 per day. 

Technical Sessions 

An attractive program of technical papers and presentations is being arranged 
by the Papers Committee, laying special emphasis upon the developments in the 
technic, equipment, and practices of the studios. Several sessions will be held in 
the evening, to permit those to attend who would be otherwise engaged in the 
daytime. All sessions will be held at the Hotel. 

Studio and Equipment Exhibit 

The exhibit at this Convention will feature apparatus and equipment developed 
in the studios, in addition to the usual commercial equipment. All studios are 
urged to participate by exhibiting any particular equipment or devices they may 
have constructed or devised to suit their individual problems, conform to their 
particular operating conditions, or to achieve economies in production, facilitate 
their work, or improve their products. 

Those desiring to participate should communicate with the General Office of 
the Society, Hotel Pennsylvania, New York, N. Y. No charge will be made for 
space. Each exhibitor should display a card carrying the name of the particular 
studio or manufacturer, and each piece of equipment should be plainly labelled. 
In addition, an expert should be in attendance who is capable of explaining the 
technical features of the exhibit to the Convention delegates. Expenses inci- 
dental to installing and removing equipment, and the cost of any power consumed, 
will be borne by the exhibitors. 

Semi-Annual Banquet 

The semi-annual banquet of the Society will be held at the Hotel on Wednesday, 
May 22. Addresses will be delivered by prominent members of the industry, 
followed by dancing and entertainment. Tables reserved for 8, 10, or 12 persons ; 
tickets obtainable at the registration desk. 



374 SPRING, 1935, CONVENTION [J. S. M. P. E. 

Studio Visits 

S. M. P. E. delegates to the Convention have been courteously granted the 
privilege of visiting and inspecting the Warner Bros. First National Studio 
(courtesy of the Electrical Dept.), the Fox Hill Studio of Fox Film Corp., and the 
Walt Disney Studio: admission by registration card only. A visit has also been 
arranged to the California Institute of Technology. All bus charges on studio and 
other trips will be assumed by the individual delegates. 

Motion Pictures 

Passes will be available during the Convention to those registering, to Grau- 
man's Chinese and Egyptian Theaters, Pantages', Hollywood Theater, Warner 
Bros.' Hollywood Theater, and Gore Bros.' Iris Theater. 

Ladies' Program 

An especially attractive program for the ladies attending the Convention is 
being arranged by Mrs. E. Huse, hostess, and her Ladies' Committee. A suite 
will be provided in the Hotel where the ladies will register and meet for the various 
events upon their program. 

Further details of the Convention will be published in the next issue of the 
JOURNAL. 

Points of Interest 

En route: the gigantic Boulder Dam project, Las Vegas, Nevada; Union 
Pacific Railroad. 

Hollywood and vicinity: Beautiful Catalina Island; Zeiss Planetarium (Open 
May 1); Mt. Wilson Observatory; Lookout Point, on Lookout Mountain; 
Huntington Library and Art Gallery (by appointment only) ; Palm Springs, Calif. ; 
beaches at Ocean Park and Venice, Calif.; famous old Spanish missions; Los 
Angeles Museum (housing the S. M. P. E. motion picture exhibit) ; Mexican village 
and street, Los Angeles; the California Pacific International Exposition at San 
Diego, Calif, (open May 29); Agua Calienta, Mexico; Tia Juana, Mexico. 

Golf 

Members attending the Convention have been extended the privileges, at the 
usual course rates, of the following courses: 
Hollywood Country Club, North Hollywood 
Oakmont Country Club, Glendale 
Westwood Country Club, West wood 
Rancho Golf Club, Westwood 

PROGRAM 
Monday, May 20 
9:00 A.M. Florentine Room 
Registration 

10:00 P.M. Society Business 

Technical Papers Program 



April, 1935] 



SPRING, 1935, CONVENTION 



375 



12:30 P.M. 



2:00 P.M. 



8:00 P.M. 



New Supper Room 

Informal Get-Together Luncheon, for members, their families, 
and guests; short addresses by eminent members of the indus- 
try. Tickets obtainable at registration desk. 

Florentine Room 
Technical Papers Program 

Visit to Walt Disney Studio. 

Direction of Mr. W. Garity, Studio Manager; admission by reg- 
istration card only; buses leave the Hotel promptly at 
7:30 P.M. 



Tuesday, May 21 

9:30 A.M. Florentine Room 

Technical Papers Program 

1:30 P.M. Visit to Warner Bros. First National Studio. 

Luncheon and inspection of studio; courtesy of the Electrical 
Department, under the direction of Mr. F. Murphy, Chief 
Studio Engineer; admission by registration card only; buses 
leave the Hotel promptly at 1 :00 P.M. 

7:30 P.M. Academy Room 

Technical Papers Program 

Wednesday, May 22 

9:30 A.M. Florentine Room 

Technical Papers Program 

2:30 P.M. Visit to Fox Hill Studio. 

Courtesy of Fox Film Corp.; direction of Mr. W. J. Quinlan, 
Chief Studio Engineer. Admission by registration card only; 
buses leave the Hotel promptly at 2:00 P.M. 

7:30 P.M. New Supper Room 

Semi- Annual Banquet of the S. M. P. E. 

Addresses by several eminent members of the industry; danc- 
ing, and entertainment; tables reserved at the registration 
desk for 8, 10, and 12 persons. 

Thursday, May 23 

9:30 A.M. Florentine Room 

Technical Papers Program 



376 SPRING, 1935, CONVENTION 

2:30 P.M. Visit to California Institute of Technology. 

Direction of Dean F. W. Hinrichs, Jr. ; inspection of astronomical, 
aeronautic, and high-voltage laboratories; admission by regis- 
tration card only; buses leave the Hotel for Pasadena promptly 
at 1:30 P.M. a beautiful scenic trip. 

7:30 P.M. Academy Room 

Technical Papers Program 

Friday, May 24 

9:30 A.M. Florentine Room 

Technical Papers Program 

2:00 P.M. Florentine Room 

Technical Papers Program 

Adjournment of the Convention 



SOCIETY ANNOUNCEMENTS 

ATLANTIC COAST SECTION 

At the regular monthly meeting, held on February 20 at the Hotel Pennsyl- 
vania, New York, N. Y., Mr. Eddie Senz delivered a talk on the subject of 
"Make-Up for Motion Pictures," discussing the subject from both the esthetic 
and the photographic points of view. Miss Dean Fureau, of Paramount, kindly 
acted as model for Mr. Senz, who illustrated the technic of his art by making up 
Miss Fureau 's face in exactly the same manner as it would be made up for actual 
motion picture production. The effects of various shadings and colorings of the 
features in order to accentuate favorable facial characteristics or to subdue un- 
favorable ones were demonstrated. Mr. W. Graff, of Hudnut's, New York, as- 
sisted by illustrating how the hair was dressed to suit the personality of the 
subject. 

More than two hundred persons attended the meeting, and an interesting dis- 
cussion followed the presentation. Lighting equipment was kindly supplied by 
Mr. M. W. Palmer. 

PROJECTION PRACTICE COMMITTEE 

The first meeting of the Committee under the new chairman, Mr. J. O. Baker, 
was held at the Paramount Building, New York, on March 6. Work on revis- 
ing the projection room layouts originally published in the August, 1931, JOURNAL, 
was continued. It is planned to present the revised specifications and layouts at 
the Spring Convention at Hollywood in May. 

PAPERS COMMITTEE 

At a meeting held at the General Office of the Society on February 21, plans were 
laid for constructing the papers program for the coming Convention at Holly- 
wood. Nine technical sessions will be held and there will be many interesting 
presentations and demonstrations. The program is under the direction of Mr. 
J. I. Crabtree, Editorial Vice-President, Mr. J. O. Baker, Chairman of the Papers 
Committee, assised in the Hollywood area by Mr. W. A. Mueller. Details of the 
Convention may be found on page 372 of this issue of the JOURNAL. 

SECTIONAL COMMITTEE ON MOTION PICTURES UNDER THE A. S. A. 

As reported previously in the JOURNAL, a Sectional Committee on Motion Pic- 
tures, organized according to the procedure of the American Standards Associa- 
tion, is in the process of being formed, with the S. M. P. E. as sponsor. The ap- 
pointments thus far made by the various organizations and societies are as 
follows: 

377 



378 



SOCIETY ANNOUNCEMENTS 



[J. S. M. P. E. 



American National Committee for International Congresses 

of Photography W. Clark 

Acoustical Society of America F. L. Hunt 

Illuminating Engineering Society R. E. Farnham 

Fire Protection Group of the A. S. A. A. R. Small 

Amateur Cinema League F. G. Beach 

Eastman Kodak Company L. A. Jones 

Agfa Ansco Corp. P. Arnold 

Dupont Film Mfg. Corp. N. F. Oakley 

National Electric Mfr's. Assoc. J. G. T. Gilmour 

Theater Equipment Supply Mfr's. Assoc. O. F. Neu 

Bell & Howell Co. J. A. Dubray 

Akeley Camera Co. J. L. Spence 

Mitchell Camera Co. G. H. Worrall 

Electric Research Products, Inc. C. Flannagan 

RCA Manufacturing Co. M. C. Batsel 

International Projector Corp. H. Griffin 

National Carbon Co. W. C. Kunzmann 

Motion Picture Producers & Distributors of America A. S. Dickinson 

D. Palfrey man 

U. S. Bureau of Standards E. W. Ely 

Bell Telephone Laboratories, Inc. H. M. Stoller 

Independent Supply Dealers' Assoc. J. E. Robin 

Nine organizations have yet to designate their representatives, including the 
Society of Motion Picture Engineers. Additions to the list will be published in a 
subsequent issue of the JOURNAL. When complete, the list is to be approved by 
the Council of the American Standards Association, whereupon an organization 
meeting will be called and an agenda of projects for standardization drawn up. 
The chairman of the Committee is yet to be chosen; the secretary is S. Harris, of 
the General Office of the S. M. P. E. 

PROGRESS MEDAL AWARD 

At a recent meeting of the Board of Governors, the provisions quoted below 
referring to the Progress Medal Award of the Society of Motion Picture Engineers 
and known as Section 3 of "Administrative Practices" of the Board, were adopted. 
The Committee appointed to administer these provisions for the year 1935 con- 
sists of: 

A. N. GOLDSMITH, Chairman 

M. C. BATSEL C. DREHER 
J. CRABTREE W. B. RAYTON 

The design of the medal, recently approved by the Board, was the work of 
Mr. Alexander Murray, of Rochester, N. Y., and was generously donated by him 
to the Society. 

A meeting of the Committee will be held on June 27, allowing sufficient time 
to determine the recipient of the Award before the Fall, 1935, Convention, at 
which time the award will be made. All Fellows and Active members of the 
Society are urged to consider this matter, and, if they have any concrete proposals, 



April. 1935] SOCIETY ANNOUNCEMENTS 379 

to forward them to the General Office of the Society at the earliest convenient 
date. A description of the medal follows : 

Obverse: The center is a replica of the official emblem of the Society. Above 
and around the emblem are embossed the words "For Progress," and below are two 
laurel branches, Grecian symbols of achievement. A reproduction of film per- 
forations forms a decorative motif surrounding the central portion of the design. 
Eleven concave panels fill the remaining area extending to the outer edge of the 
face, upon each of which the form of a bird in flight is embossed. Various move- 
ments of the flight are depicted, reproducing the work of E. Marey, a French 
scientist who, in 1886, designed a "photographic gun," using circular glass plates, 
for analyzing the movements of living things. Although it was not Marey's in- 
tention to reproduce motion, his plates embodied the essential elements of the 
motion picture and the representation of them is therefore symbolic of the early 
development of motion pictures. 

Reverse: The central portion consists of a series of horizontal oblong panels 




(Obverse) (Reverse) 

The Progress Medal 

arranged in a partial pyramidal form and bearing the embossed inscription 
"Awarded to (Name of Medalist) for Outstanding Achievement in Motion Pic- 
ture Technology." Crystals of silver bromide, the light-sensitive salt used in 
most photographic emulsions, are reproduced in two of the panels. Below the 
inscription is engraved the year of the award. Above it is a small rectangular 
panel upon which is engraved a sensitometric curve, representing the classical 
researches of Hurter and Driffield, who laid down much of the fundamental theory 
regarding numerical specification of photographic emulsion characteristics. Sine 
waves, symbolic of sound and light, are embossed upon two curved panels to the 
left and right of the central pyramid. In a slightly inclined panel surrounding al- 
most the entire outer edge, the name of the Society appears in embossed letters. 

REGULATIONS 

A medal shall be awarded in the year 1934, and may be awarded in subsequent 
years, to be known as the Progress Medal, and the sum of $185 is appropriated 



380 SOCIETY ANNOUNCEMENTS 

for the design and die of this medal. The medal shall be awarded to an individual 
in recognition of any invention, research, or development which in the opinion of 
the Progress Award Committee shall have resulted in a significant advance in the 
development of motion picture technology. The Progress Award Committee 
shall consist of not less than five Fellows or Active members of the Society, to be 
appointed by the President subject to ratification by the Board of Governors. 
The names of the persons deemed worthy of the award may be proposed and 
seconded in writing by any two Fellows or Active Members of the Society, and 
shall be considered by the Committee not later than the month of July. A 
written statement of the accomplishments shall accompany each proposal. 

Notice of the meeting of the Progress Award Committee must appear not later 
than the June issue of the JOURNAL. All proposals shall reach the chairman not 
later than June 20. 

A majority vote of the entire Committee shall be required to constitute an 
award of the Progress Medal. Absent members may vote in writing. The re- 
port of the Committee shall be presented to the Board of Governors for ratification 
at least one month before the Fall Meeting of the Society. The recipient of the 
Progress Medal shall be asked to present a photograph and pertinent technical 
biographical data of himself to the Society and, at the discretion of the Com- 
mittee, he may be asked to prepare a paper for publication in the JOURNAL of the 
Society. These regulations, the names of those who have received the medal, 
the year of each award, and a statement of the basis for each award shall be pub- 
lished annually in the JOURNAL of the Society. 

MID-WEST SECTION 

A showing of the American Cinematographer prize-winning 16-mm. films was 
held at the March 14 meeting of the Section at the Bell & Howell auditorium, 
Chicago, 111. The meeting was well attended, and the showing aroused consid- 
able interest. Other meetings of the Section are planned for April 4 and May 2. 



JOURNAL 






OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 

Volume XXIV MAY, 1935 Number 5 



CONTENTS 

Page 
A Revolving Lens for Panoramic Pictures F. ALTMAN 383 

Recent Developments in the Acoustics of Motion Picture Sound 
Stages M. RETTINGER 395 

On the Relation between the Shape of the Projected Picture, 
the Areas of Vision, and Cinematographic Technic 

B. SCHLANGER 402 

Mechanical Recording on Film A. F. CHORINE 410 

The Educational Motion Picture of Yesterday, Today, and 
Tomorrow H. A. GRAY 414 

The Theatergoer's Reaction to the Audible Picture as It Was 
and Now M. HALL 424 

A Glossary of Color Photography 432 

New Motion Picture Apparatus "... 450 

Officers and Governors of the S. M. P. E 457 

Committees of the S. M. P. E 459 

Society Announcements 463 

The Spring, 1935, Convention at Hollywood, Calif .; Tentative 
Papers Program 465 



JOURNAL 

OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 



SYLVAN HARRIS, EDITOR 

Board of Editors 
J. I. CRABTREE, Chairman 

O. M. GLUNT A. C. HARDY L. A. JONES 

J. O. BAKER 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1935, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 



Officers of the Society 

President: HOMER G. TASKER, 4139 38th St., Long Island City, N. Y. 
Past-President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
Executive Vice-President: EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, 

Calif. 

Engineering Vice-President: LOYD A. JONES, Kodak Park, Rochester, N. Y. 
Editorial Vice-President: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Financial Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Convention Vice-President: WILLIAM C. KUNZMANN, Box 6087, Cleveland, Ohio. 
Secretary: JOHN H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York, N. Y. 

Governors 

MAX C. BATSEL, Front & Market Sts., Camden, N. J. 
LAWRENCE W. DAVEB, 250 W. 57th St., New York, N. Y. 
ARTHUR S. DICKINSON, 28 W. 44th St., New York, N. Y. 
HERBERT GRIFFIN, 90 Gold St., New York, N. Y. 
GERALD F. RACKETT, 823 N. Seward St., Hollywood, Calif. 
WILBUR B. RAYTON, 635 St. Paul St., Rochester. N. Y. 
SIDNEY K. WOLF. 250 W. 57th St., New York, N. Y. 



A REVOLVING LENS FOR PANORAMIC PICTURES' 



F. ALTMAN** 

Summary. The panoramic lens described was developed at a time when such a 
lens seemed a desirable means of achieving the large angle of view required under cer- 
tain conditions for the so-called wide film, the film width in question being 70 mm. 
The full-field angle across the film with this lens was nearly 50 degrees, as compared 
with about 28 degrees for standard 35-mm. film using the same focal length lens. 
The film is held in a gate curved horizontally, the curve facing the lens and of a 
radius approximately equal to the focal length of the lens. The lens revolves about its 
rear nodal center, so that the axis of the lens sweeps out the entire angle of the picture. 
Because of the design of the lens, and a special shutter revolving with it in a plane near 
the film, allowing only a limited portion of the field to be imaged upon the film at a 
given instant, a sharp picture results. 

The development of the lens that forms the subject of this paper 
was done some years ago when the interest in wide pictures made 
such a panoramic lens seem a desirable means of achieving the large 
angle of view required under certain conditions for this film. If the 
interest in a picture of the width here considered has passed, or if 
in the meantime ordinary lenses of sufficient aperture and with wider 
covering power than those then available have been developed, these 
facts may not make the present paper without some interest to the 
Society. 

At the time this work was undertaken the 35-mm. film aperture 
had been reduced to make room for the sound-record, with the result 
that for a time, at least, the old proportions that had been standard 
for years had been disturbed. A difference of opinion as to exactly 
what width of sound-track would be necessary for properly recording 
the sound was, no doubt, one of the factors that gave the advocates of 
the wider film the opportunity for which they had waited. 

Of the various widths of film considered at that time the 70-mm. 
was the one for which the present lens was developed. This film pro- 
vided for a picture area of 46 X 23 mm., and for a sound-track of 
250 mils. This picture area had a diagonal of 51.4 mm. and required 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 
** Eastman Kodak Co., Rochester, N. Y. 



384 



F. ALTMAN 



[J. S. M. P. E. 



a half -angle slightly greater than 27 degrees to cover the corner of the 
picture with a 50 mm. lens. The full-field angle across the film 
with such a lens was nearly 50 degrees, as compared with a field of 
slightly more than 28 degrees with standard 35-mm. film using the 
same focal length lens. Longer focal length lenses were available 
that would cover this new larger aperture subtending a smaller angle 
of view, but studio practice seemed to demand the shorter 50-mm. 
lens for a great portion of the work. 

Unfortunately, lenses have a limiting angle-of-field over which 
they will form a sharp image. This angle varies among different 
types, but few if any were then available that would meet the require- 




Illustrating determination of "equivalent sur- 
face of refraction." 

ments of the 46 X 23-mm. gate with a 50-mm. focal length. The 
light value of oblique pencils diminishes rather markedly at these 
extreme angles due to the trimming of the lens mounts. Both these 
factors set up limitations beyond which it is not practicable to go with 
the normal lens. Some other method of increasing the angle-of- 
view must be adopted. 

Various efforts had been made prior to this time to produce a pic- 
ture of larger angular dimensions, and some spectacular results 
had been achieved. Notable among them were the Widescope pic- 
tures, made with a two-lens two-film camera. In this attempt each 
lens recorded only half the field-of-view. These pictures were de- 
scribed and demonstrated before the Society by J. D. Elms in 1922. 1 
Still another was the panoramic pictures made with a revolving lens 
by G. C. Ziliotto 2 and described before the Society in 1924. 

The difficulties inherent in the Widescope pictures using two 
lenses and two films that must be properly synchronized in taking 
and projecting, as well as the difficulties in uniformly processing the 
two films, certainly would leave much to be desired, however satis- 



May, 1935] 



A REVOLVING LENS 



385 



factory the final results upon the screen might be. The pictures 
made with a revolving lens by Ziliotto were spectacular in character, 
but since they were taken at the rate of eight frames per second the 
process was limited to still scenes. 

Elms had continued his efforts to make a wide picture, and taking 
up the revolving lens as used by Ziliotto had attempted to increase 
the number of pictures per second by oscillating the lens. In a 
camera developed by him for this purpose, the lens was stopped at 




FIG. 2. Illustrating the action of the panoramic lens with curved 
film plane. 

the end of the exposure of one frame and the direction reversed, thus 
sweeping out each succeeding frame by rotating the lens in the op- 
posite direction. In this manner he was able to double the number 
of frames per second made by Ziliotto, but mechanical difficulties 
introduced by stopping and reversing the direction of the lens re- 
sulted in noise, vibration, and wearing of parts. 

In 1927, Elms came to the Kodak Company with the proposal that 
they develop a panoramic lens that would take a picture equally 
well with the lens in a direct or reversed direction. This problem 
was taken up at the Hawk Eye Lens plant of the company, and the 
lens that is to be described was developed as a possible solution of the 
problem. 

Lest there be any confusion as to the method of taking panoramic 
pictures in the manner here considered, let it be stated (a) that the 
film is held in a gate curved in the horizontal plane, the curve facing 



386 F. ALTMAN [j. s. M. p. E. 

the lens and of a radius approximately equal to the focal length 
of the lens; (b) that the lens is revolved about its rear nodal center so 
that the axis of the lens sweeps out the entire angle of the picture; 
and (c) that because of the special design of the lens, and because of 
a special shutter revolving with the lens and in a plane near the film, 
allowing only a limited portion of the field of view to be imaged upon 
the film at a given instant of time, a sharp picture results. In this 
manner, the angle-of-view may be greatly increased over that pos- 
sible with the orthodox type of lens. It is apparent also that the 
lens at all times is forming the picture with axial or near axial defini- 
tion and with a full and unvignetted cone of light, so that definition 
and illumination are uniform over the entire picture. 

Since the covering power required is then limited to an area de- 
fined by the height of the film and by the width of the exposure aper- 



Gmte. 




FIG. 3. The final form of the lens. 

ture (the later dimension being usually less than the height of the 
film), the demands for critical definition over this area are quite 
easily met in designing the lens. One exception was the correction 
for coma, which, because of the symmetry of the lens, proved trouble- 
some. The solution of this problem lead to a rather unconventional 
design. 

It might be well at this point to mention the several requirements 
that are to be fulfilled in designing a panoramic lens. We speak of 
a lens having a certain focal length, and some very carefully use the 
term "equivalent" focus without perhaps visualizing exactly what the 
term implies. In the case of a simple lens we would make no serious 
error in stating that the focal length was the distance from the lens 
to the point where light from a distant object was brought to focus. 
In the case of the complicated structure to which the modern anastig- 
mat has so often developed, this surface-to-image distance would in 
most cases be in error. The opticians tell us that if we project paral- 
lel rays of light through a lens as though they were undeviated by the 



May, 1935) 



A REVOLVING LENS 



387 



lens, until they intersect the extensions of the final refracted rays 
emerging from the lens, the intersections of the common pairs (the 
refracted and the unrefracted ray) will define a surface called an equiva- 
lent surface of refraction. (See Fig. 1.) It is as though the rays 
of light had continued through the lens undeviated until they en- 
countered this surface, and that 
all refraction had then taken 
place at this point. If we were 
to reverse the lens and consider 
the same parallel rays entering 
the opposite end of the lens, a 
similar equivalent surface of re- 
fraction would be found. 

The points where these 
equivalent refracting surfaces 
cut the axis of the lens locate 
the position of the nodal points 
or principal planes that are im- 
portant in our present discus- 
sion. These principal planes as 
determined by the position of 
the nodal points are the basis 
of our calculation of focal 
length, object and image dis- 
tance, and magnification. These 
nodal points have another im- 
portant property: If a ray of 
light is directed to the first 
nodal point, it will leave the 
lens as though it came from 
the second nodal point, and 
the ray will be characterized 
further by having the same direction as it had before entering the 
lens. It is obvious, then, that if we revolve a lens about this rear 
nodal point, there will be no lateral shift of the image for any 
amount of rotation, provided the lens is free from distortion. 

Thus far we have recorded two important requirements for a 
rotating lens. We must revolve it about the rear nodal center and 
the lens must be free from distortion. If we are to revolve the lens 
completely, however, and attempt to take a picture first with the lens 




FIG. 4. Early model of four-way lens in 
mounting. 



388 F. ALTMAN [j. s. M. p. E. 

in a direct and then in a reversed position, the conditions for freedom 
of lateral shift of the image requires that the front nodal point of the 
lens be also in the axis of rotation. It is, therefore, necessary that the 
two nodal centers be coincident and in the axis of rotation. Though 
it is not necessary that the lens be symmetrical, a consideration that 
succeeding pictures be identical in quality would lead to symmetry 
in design. 

As previously mentioned, in taking a panoramic picture the film 
must be curved toward the rotation center of the lens with a radius 
of curvature approximately equal to the focal length of the lens. 
Before considering the design of the lens further, it might be well to 
visualize just what happens when we make a panoramic picture. In 
the vertical direction the film is flat, being curved only in the hori- 
zontal direction. Now if the field of the lens is flat and the film so 
curved, the image plane will be tangent to the film at succeeding 
points during exposure (Fig. 2) ; and if the exposure were limited by 
a very small slit, definition would be as perfect as though the image 
were formed upon a flat film. The requirements for exposure, how- 
ever, necessitate a sizable width of exposure aperture, as indicated 
in Fig. 2. The size of the "out-of -focus" image resulting from the use 
of various shutter apertures, as well as the corresponding exposure 
obtained, are shown in Table I. 

TABLE I 

Exposure and Maximum Circle of Confusion with Various Shutter Openings 

(Two-way lens ; aperture ratio, f/3. 5; 12 revolutions per sec. ; 24 frames per sec. ; 

shutter radius, 40 mm.) 



Slit 
Size 
(mm.) 


Angle of 
Slit 
(degrees) 


Max. "Out 
of Focus" 
(mm.) 


Circle of 
Confusion 
(mm.) 


Circle of 
Confusion 
(inches) 


Expo- 
sure 
Time 


5 


6 


0.06 


0.017 


0.0006 


1/720 


10 


12 


0.27 


0.077 


0.0030 


1/362 


12 


14 


0.39 


0.110 


0.0040 


1/300 


15 


21 


0.87 


0.250 


0.0100 


1/200 


20 


28 


1.61 


0.46 


0.0180 


1/150 



In taking 24 frames a second with a two-way lens, the speed of 
rotation of the lens will be twelve per second. If we use an aperture 
in the shutter subtending 6 degrees, the exposure time will be Veo th 
of the total, or Vrao sec. The table shows that the image due to the 
curvature of the film will be only 0.06 mm. out of focus at the begin- 
ning of the exposure of any given point; exactly in focus at the middle 



May, 1935] 



A REVOLVING LENS 



389 




of the exposure time, when the image is formed upon the axis of the 
lens; and that it will be again out of focus by 0.06 mm. at the end of 
the exposure. If the lens is working at//3.5, the circle of confusion 
on the film caused by these out-of-focus images will be 0.017 mm., or 
about 0.0006 inch. 

With a slit subtending 12 degrees, the exposure time is Vseo sec., 
and the maximum circle of confusion of the out-of-focus image on the 
film will be 0.077 mm., or about 0.003 
inch. We see that the increase of expo- 
sure by using a wider exposure aperture 
is achieved with a lowering of the defi- 
nition. Definition is quite good even 
with larger apertures than those con- 
sidered; because a large portion of the 
exposure, even in the case of wider 
apertures, is made with image sizes that 
constitute critical definition, and only a 
fraction of the exposure is made with 
"out-of-focus images" approaching soft 
definition. In practice, some compro- 
mise is made in the curvature of the 
gate which tends to reduce the maxi- 
mum departure of the image from the focal plane of the film. 

Since, however, there is this deterioration of the image as the 
shutter aperture increases varying, as it does, with the cosine of half 
the angle subtended by the shutter aperture it would appear that for 
a fixed rotation speed, the exposure might best be increased by in- 
creasing the lens aperture; for the reason that the "out-of-focus 
image" increases in size only in proportion to the diameter of the lens, 
whereas the exposure increases as the square of this value for a given 
focal length. Thus, if we make the lens//2.3 instead of //3.3, the ex- 
posure time will be increased by 100 per cent, whereas the out-of- 
focus image size would increase by only 40 per cent. 

Another method of increasing the exposure without the attendant 
loss of definition presented itself and was successfully applied in 
designing this lens. If the lens system were so designed that the space 
between the front and the rear components was sufficient to allow an 
identical system to be placed in the same plane with its axis at right 
angles to the first, then four pictures could be made during each revo- 
lution, and the lens would revolve only six times per second to take 



FIG. 5. Method of mounting 
revolving lens. 



390 



F. ALTMAN 



[J. S. M. P. E. 



twenty-four pictures; and with a constant exposure aperture, the 
exposure time would be increased 100 per cent over that of the two- 
way lens without any loss in definition. 

This, then, was the course decided upon, and a symmetrical doublet 
having a central separation sufficient for mounting the lenses as de- 
scribed above was developed. As to the various aberrations that 
require correction in a lens system, they will merely be mentioned 





FIG. 6. Panoramic camera developed by J. D. Elms. 

here. Color error on the axis and color difference of magnification 
depend for correction upon the powers of the elements, their positions 
with respect to each other in the system, and upon the dispersive 
values of the glasses used. Spherical aberration and astigmatism 
are corrected by altering the shapes of the various elements. Dis- 
tortion, usually responsive to the distribution of power of the several 
elements, was not a factor in this lens, due to its symmetry. Field 
curvature is quite completely controlled by suitably choosing the 
glasses and the powers of the several components. Only one correc- 
tion proved troublesome: the correction for coma. 



May, 1935] 



A REVOLVING LENS 



391 



Coma is manifest as a one-side blurring of oblique images, and is the 
result of spherical aberration of the oblique pencils of light. Coma 
of the character here considered is corrected by fulfilling certain con- 
ditions in the design of the lens into which we shall not go at this 
time. Let it be sufficient to say that this condition is satisfied in a 
symmetrical lens such as we have developed when, and only when, 
such a lens is working at unit magnification, when object and image 
are at equal distances from the lens. When such a symmetrical lens is 
used to image an object at a great distance, however, the condition for 
absence of coma no longer obtains, and to fulfill the necessary con- 
ditions for its correction some de- 
parture from symmetry must be 
made. 

To render the lens system un- 
symmetrical so that this condition 
might be met, and yet provide for 
rotating the lens, the following ex- 
pedient was adopted. To the 
symmetrical rotatable doublet al- 
ready described was added a fixed 
or stationary member. This latter 
consisted of a negative meniscus 
element with surfaces concentric 
with the nodal center or axis of 
rotation of the rotatable doublet. 
Thus was achieved the lack of 
symmetry in the system as a 
whole which enabled the sine con- 
dition to be fulfilled, with conse- 
quent correction of the coma 
without disturbing the conditions 
necessary for rotatability pre- 
viously established. The negative lens, having radii concentric 
about the point of rotation, would have not one optical axis 
as does an ordinary centered optical element, but innumerable axes, 
comprising all lines drawn through their common center of curva- 
ture. It is therefore seen that, regardless of the position of the sym-. 
metrical system in a plane containing its optical center and the optical 
center of the negative meniscus component, the complete lens would 
constitute a centered system. Both nodal points of the negative 




FIG. 7. Clipping of picture 
made under normal studio condi- 
tions, at 24 frames per second. 



392 F. ALTMAN [J. S. M. P. E. 

meniscus with concentric radii are at one and the same point, which is 
their common center of curvature; and since this point is made com- 
mon with the nodal center of the symmetrical component, there is 
no shift of this nodal center when the negative is added to the system. 
The only effect is to increase the focal length of the system. Since 
in this case we are limited mechanically in the size of the lenses con- 
stituting the symmetrical components of the system, this resulted in 
reducing the effective aperture to //3.3. 

The correction of coma by the addition of the negative meniscus 
might be explained as a result of two definite conditions that now 
obtain. First, the negative lens forms a virtual image of a distant 
image at its own focal distance in front of the symmetrical component. 
This distance approximates so nearly the conditions for unit magni- 
fication for the symmetrical portion of the lens that this virtual image 
is imaged by the latter free from coma. This, in fact, was the ap- 
proach to the solution, and the results seem to bear out the reason- 
ing. The aberration and sine condition are well corrected in the final 
form of the lens, which is shown in Fig. 3. 

It was necessary only to give to the stationary lens sufficient aper- 
ture to assure a full cone of light to the rotating member over its en- 
tire travel; and then, after the lens was constructed, to make some 
rather careful adjustments of the mounting. Looking after such de- 
tails as to make all components in any particular case of the same 
piece of glass, and resorting to a special method of grinding the large 
meniscus, which obviously could cause trouble in centering, completes 
the story. Figs. 4 and 5 illustrate the revolving lens, and Fig. 6 a 
panoramic camera developed by Elms. Fig. 7 shows a clipping of 
a picture made under normal studio conditions, at a speed of 24 
frames per second. 

REFERENCES 

1 ELMS, J. D.: "Demonstration and Description of the Widescope Camera," 
Trans. Soc. Mot. Pict. Eng., VI (1922), No. 15, p. 124. 

2 ZILIOTTO, G. C. : "Panoramic Motion Pictures," Trans. Soc. Mot. Pict. 
Eng., VIII (1924), No. 18, p. 206. 

DISCUSSION 

MR. MITCHELL : In the illustration you showed, you had a curved line on the 
right-hand side. 

MR. ALTMAN: That was the curved aperture. 

MR. MITCHELL: What was the size of the lens used for photographing the 
motion picture frame we saw? 



May, 1935] A REVOLVING LENS 393 

MR. ALTMAN: That was a 50-millimeter focal length lens. 

MR. MITCHELL: What was its physical size? 

MR. ALTMAN: The overall length of the symmetrical component was 25.42 
mm. The elements of this part of the lens were 15.7 mm. in diameter. The 
stationary negative element was 12 mm. thick, and the outer and inner curves 
were 30.00 and 18.00 mm., respectively. The element was spaced 5.29 mm. 
from the rotatable member or 18.00 mm. from the center of the lens system. The 
stationary element was 43.00 mm. in diameter. 

CHAIRMAN CRABTREE : What does the lens weigh, and how is it supported for 
rotating? 

MR. ALTMAN: It was very light. The rotating part has very little mass. It 
was mounted upon a shaft so that the adjustments we spoke of could be made. 
Of course, the two systems had to be perfectly aligned so that the optical centers 
were not only together but on the axis of rotation in both cases ; and there could 
not be any elevation difference, so that one picture would be high and the next 
low. To take care of that we had to have a jig made and examined on the lens 
bench with the microscope, to make sure that the four images came around right 
on the cross-wire and there was no shift laterally or vertically. 

MR. McGuiRE : Does Mr. Elms know where wide pictures are now being shown? 

MR. ELMS: I think Mr. Spoor showed a few wide pictures at the World's 
Fair last year. 

MR. McGuiRE: Mr. Elms may be pleased to learn that yesterday I escorted 
a group of our members through the factory of the International Projector Corp. 
and showed wide pictures with one of the special Grandeur projectors. Six of 
them were specially made some years ago at a cost of more than $200,000, and 
have been lying in the warehouse, unused, until yesterday. The showing of 
Grandeur pictures in the Roxy Theater is a small section of motion picture history 
and it was a great disappointment to many of us when it was decided that the in- 
dustry was not yet ready for wide pictures. I believe I am justified in stating that 
all our members who witnessed the showing of the old Grandeur pictures yester- 
day were greatly pleased, and realized that certain results could be attained with 
70-mm. which could not be achieved with 35-mm. films. Of course, it is not for me 
to say that 70-mm., 65-mm., or some other width is best, but it is still my opinion 
that wide films have certain very definite advantages. The most practical il- 
lustration of the possibilities of wide pictures can be seen in viewing of a base- 
ball game. The complete diamond can be shown and the images of the players 
are large enough to stand out very clearly. 

For nearly fifteen years we have followed Mr. Elms' work, and our factory has 
taken an active part in developing equipment through the various stages that led 
to the manufacture of the Grandeur projectors to which I have referred, and I 
believe that the public finds an additional satisfaction in seeing pictures shown in 
this manner. There are a number of us who believe in the great possibilities of 
wide pictures. It is our hope that the motion picture industry will some time 
see its way clear to adopt wide pictures as it did sound pictures. 

CHAIRMAN CRABTREE: I believe we all agree with Mr. McGuire, that for 
panoramic subjects, there is no doubt that the wide film gives a more realistic 
effect. The question is whether the additional realism given to the picture justi- 
fies the expense involved. 



394 F. ALTMAN 

Can you tell us more about the method of suspending the lens? It would seem 
to be a rather delicate job to suspend it so that the successive images would not 
move laterally or vertically. 

MR. ELMS: There is a shaft that runs to the center, top, and bottom, which is 
worked by gears. It runs very smoothly, with absolutely no vibration or noise 
of any kind. The lens that we have, I believe, weighs in the neighborhood of 
from four to six ounces, including the mounting. 

MR. ELMS, JR. : Since the question of mounting the lens has been brought up 
I would like to say that the lens was originally mounted at the Hawk-Eye Plant 
by Mr. Altman and since then has never been touched. 

There can be no lens shift since the elements are permanently locked and sealed 
in the mount, forming a complete unit and thus making it impossible for them to 
get out of alignment. The lens is encased in an inaccessible shell, since no ad- 
justments are necessary after once being mounted and sealed. 



RECENT DEVELOPMENTS IN THE ACOUSTICS OF MOTION 
PICTURE SOUND STAGES* 

M. RETTINGER** 



Summary. Methods of construction for achieving desirable reverberation char- 
acteristics in motion picture recording studios are dealt with. Ways and means are 
discussed which make possible a high degree of absorption of the low frequencies and a 
lesser degree of absorption for the high frequencies . Attention is called to the problem 
of monaural hearing; to the psychological effect of over-damped rooms; and to the 
question of feasible sound-insulation. 

Although several successful demonstrations have been given of the 
reproduction of sound with auditory perspective, the problem of re- 
cording and reproducing sound for motion pictures still remains es- 
sentially one of satisfying but one ear. Since in binaural hearing the 
attention can be concentrated upon sounds coming from a certain 
direction to the exclusion of sounds coming from other directions, the 
reverberation of the sounds and noises coming from different places 
can to a high degree be suppressed when listening with both ears. 
In monaural hearing, however, in which this faculty for listening to 
sounds coming from a single direction and ignoring sounds from all 
other directions is very much weakened, it is necessary that the room 
have a lower period of reverberation in addition to a lower level of 
noise than would be required for satisfactory hearing with two ears. 

Hence it has become well-known practice in motion picture studio 
design to make the walls and ceiling of the stages as absorptive as 
possible. By doing so, the acoustical quality of the sound recording 
will be determined largely by the set materials and dimensions, thus 
permitting a greater freedom and simplicity in the acoustical design 
of the sets. It is the general practice to treat the inner walls of the 
sound stages with a 4-inch fill of mineral or rock wool between 2X4- 
inch wood studs covered with cloth or wire screen, and to treat the 
ceiling with a iy 2 - to 2-inch rock or mineral wool blanket. With 

* Received January 2, 1935. 
** Pacific Insulation Co., Los Angeles, Calif. 

395 



396 



M. RETTINGER 



[J. S. M. P. E. 



such a treatment a stage of volume, say, 600,000 cubic feet, will have 
a reverberation period of about 1 second at 128 cycles per second and 
of about 0.8 second at 512 to 2048 cycles per second. This condition 
is shown graphically in Fig. 1. 

Fig. 2 shows the absorption characteristic of granulated rock wool, 
6 inches thick, 12 Ib. per cubic foot, filled between 2 X 6-inch wood 
studs, 16 inch o.c., covered with cheesecloth, as determined by Knud- 



sen. 



From Figs. 1 and 2 it is obvious that the established criteria for 
reverberation and absorption characteristics pertaining to ordinary 
speech and music rooms do not hold true for sound stages, 2 although 



1.0 



ii - 

tf z 

uJ 

cr 



118 



IOZ-1- -2.0-48 



2,56 5I7. 

FREQUENCY 

FIG. 1. Reverberation period of sets treated with 4-inch 
fill of mineral or rock wool between 2 X 4-inch wood studs 
covered with cloth or wire screen; ceiling with y 2 - to 2-inch 
rock or mineral wool. 

reverberation controllers, (i. e., variable absorbents in the shape of 
large panels covered on one side with a highly absorptive material and 
on the other side with a reflective one) are installed in many stages to 
allow some alteration of the reverberation time for different musical 
performances in accordance with the known principles of architectural 
acoustics. 

As it would be beyond the scope of this paper to describe the actual 
conditions in the many sound stages that the author has investigated, 
only very general principles recently developed for absorption control 
in highly damped rooms can be mentioned. A notable contribution 
of this kind is due to G. von Be'kse'y, 3 and deals with the elimination 
of the generally greater absorption at the high frequencies. 4 When 



May, 1935] 



ACOUSTICS OF SOUND STAGES 



397 



sound is recorded in a room having excessive high-frequency absorp- 
tion, the reproduced speech loses naturalness and the reproduced 
music brilliance. High frequencies display such decidedly limited 
directional qualities that, with the walls highly absorbent for them, 
the energy of these frequencies can not uniformly distribute itself 
throughout the room. Fig. 3 shows the directional characteristics 
of a tone having a frequency of 2000 cycles per second. The solid 
curve represents the intensity at points on a circle out in the open 
at the angles indicated on the abscissas. The dotted curve shows 
that when in an 1100-cubic-meter room the time of reverberation is 
1.0 second for the 2000-cycle tone, there still exists, due to the direc- 



Z .8 

o y - 



t - 6 
s- 5 

V - 

i- 

t * 
g 



IZ8 



Z56 5lt IOZ4- Z048 4096 
*' rRC.QUE.NCV 

FIG. 2. Absorption characteristic of granulated rock 
wool, 6 inches thick, 12 Ib. per cu. ft., between 2 X 6-inch 
wood studs covered with cheese cloth. 

tional effect, a lack of intensity at the previously mentioned points. 
The curve made of dots and dashes, however, which shows the inten- 
sity at the same points in the same room when the time of reverbera- 
tion is 1.6 seconds for the 2000-cycle tone, is flat enough to allow a 
uniform distribution of sound energy throughout the room, thus per- 
mitting good recording from almost any point in the sound stage. 

Hence, acoustical materials having decreased absorption at the 
higher frequencies are frequently demanded. If it is further taken 
into consideration, however, that frequencies above 2000 cycles per 
second are to a great extent absorbed by the air, 5 it becomes a prob- 
lem to find a really satisfactory absorbing material for sound stages, 
that is, a material with a sufficiently small absorption coefficient for 
the high frequencies. Realizing this, von Bekesy searched for such 



398 M. RETTINGER [j. s. M. P. E. 

a material, although he chose the alternative; that is, he desired a 
material having a high degree of absorption for the low frequencies. 
Fig. 4 shows the absorption curve for his material. It is made up of 
moderately tightly stretched canvas behind which is a 4-cm. layer of 
very loose cotton. The canvas must not be stretched too tightly, be- 
cause it must be allowed to vibrate as a membrane. 

Another way to increase the absorption at a certain frequency* 
consists in introducing an air-space between the absorptive material 
and the rigid wall of the studio. This air-space should be equal to 
about one-fourth the wavelength of the tone that it is desired to ab- 
sorb to a very appreciable extent. Obviously, frequencies of slightly 
different wavelength will be absorbed highly, too, since sharp peaks 



db. 
O 

-10 



-30 



T= 1.6 




-90-6O-30 O 30 60 90 

FIG. 3. Illustrating directional qualities of 
2000-cycle tone. 



in the absorption curve of any acoustical material or construction are 
rare. Hence a considerable range of frequencies may be absorbed. 
Thus, if a movable wall of absorptive material were constructed in a 
sound stage, the absorption could be controlled nicely for almost all 
small "bands" within the audible frequency range. While such con- 
struction has, so far as this author knows, not as yet been adopted, 
a recently built broadcasting studio at Hamburg has a wall that can 
be moved by an elaborate mechanism so as to increase or decrease the 
volume of the studio, thus permitting a change in the reverberation 
time for different performances. 6 

Yet another way to increase the absorption as well as the diffusion 
of sound in a studio consists in using absorbents of different charac- 
teristics, supported independently in the form of corrugations, or 
triangular or trapezoidal "flutes." The sides of these projecting 

* Developed in the Physics Department of the University of Oslo, Norway. 



May, 1935] 



ACOUSTICS OF SOUND STAGES 



399 



forms should be comparable to the average wavelength of the sounds, 
the minimum depth being 18 inches. Such construction, besides 
providing additional area, also causes increased absorption by inter- 
nal reflection within the air-space in the back of the material, pro- 
vided, of course, that the frame is stiff enough to minimize any reso- 
nance phenomena and the normally resulting transmission of sound 
by the vibration of the structure as a whole. The plans for the Cen- 
tral Film Studios to be built at Wembley, England, specify that the 
walls of the stages have such triangular "flutes," with braces made of 
L-irons 2 x /2 X 2 1 /* X J A inch. 7 Fig. 5 shows cross-sections of two 
such possible corrugate constructions. 



L 8 
bi 

-T 

u 

1 .6 
1 9 

a .* 
tf 




w* 



80 100 



100 300 50O 1000 ZOOO 300O 
FREQUENCY 



vooo 



FIG. 4. Absorption curve of canvas stretched moderately 
tightly, backed by a 4-cm. layer of loose cotton. 



In order to minimize the fire risk if an air space is interposed be- 
tween the acoustical material and the wall, the acoustical material 
should be painted on the air-space side with a fire-resisting acoustical 
paint that will not fill up the pores, which are so necessary for the 
absorption by internal reflection. Fire-resisting screening in the 
form of perforated metal trays or tiles have for a considerable time 
been on the market, and the absorption coefficient of such materials is 
but slightly less than that of the uncovered material. 

It should be pointed out here that the absorption in a studio must 
not be allowed to become too great. When open-air conditions are 
approached in a sound stage, the reproduced music or speech becomes 
unpleasantly "dead" or "flat"; and the musicians find it difficult in 
highly padded rooms to elicit from their instruments music of the 



400 



M. RETTINGER 



[J. S. M. P. E. 



proper intensity and tone. A musician, as is aptly remarked by 
Bagenal and Wood, 8 responds especially to the sense of "power" 
afforded by a good room. Music is not one absolute tone after an- 
other, but as equence of tone relationships modified at every point by 





ABSORPTIVE. ABSORPTIVE. 

MATERIAL MATERIAL 

FIG. 5. Cross-sections of two possible corrugate 
constructions. 

the player and the room together. Some musicians have strong prefer- 
ences for certain rooms where they perform, due to the fact that the all- 
but-imperceptible tone adjustments that constitute musical "color," 
"depth," and "personality" are always influenced by room acoustics. 
Nevertheless, highly absorptive materials are much in demand by 

db. 

60 
50 
4-0 
30 

to 

10 



-1 -I O + 1 +-Z. 

LOGARITHM OF WEIGHT OT 
PANEL. PER SQ. FT. 

FIG. 6. Sound-insulation properties of homo- 
geneous partitions. 

the studios. In a paper 9 published by the author are set forth the 
conditions necessary for a material to absorb a given frequency com- 
pletely. If the experiments to be conducted at the University of 
California at Los Angeles this year bear out the calculations, the 
quality of reproduced music and speech may be enhanced even 
more than it has been in the past months by the introduction of newly 
developed electrical apparatus. 

A matter no less important than achieving the proper degree of 
absorption in a studio is to have sufficient sound-insulation to pre- 
vent the transmission of sound either from without or from within. 



May, 1935] ACOUSTICS OF SOUND STAGES 401 

The work that has been done in that direction is enormous, and again 
only very general ideas recently worked out can be mentioned. It 
has been found that the insulation of homogeneous partitions is 
proportional to the logarithm of the mass per square foot, a condition 
illustrated by Fig. 6 ; while the insulation of porous-flexible materials 
is proportional to the thickness of the material. Thus, if the thick- 
ness of a porous-flexible material such as hair-felt is increased ten- 
fold, the sound-insulation will increase ten-fold ; whereas if the thick- 
ness of a rigid non-porous material, such as brick, is increased that 
many fold, the sound-insulation will increase only two-fold. Note, 
however, that whereas a solid 9-inch brick wall has an insulation of 
50 db., two separate 4V2-inch brick walls, separated by a small air 
space, have a combined insulation of 90 db. Although a 90-db. insu- 
lation represents a very high transmission loss, structures with even 
still higher transmission loss have been constructed successfully. 
Thus the NBC studios of Radio City have in part an insulation of 
100 db. a transmission loss so great that as yet no suitable appara- 
tus is available to measure accurately such attenuation with sound 
levels normally attained in studios. 10 

The author wishes to express his sincere appreciation to Professor 
V. O. Knudsen, and Messrs. Townsend and Hansen of the Fox Film 
Corporation, for their assistance in the preparation of this paper. 

REFERENCES 

1 KNUDSEN, V. O.: "Architectural Acoustics," Wiley & Sons (New York), 
1932, p. 210. 

2 RETTINGER, M.: "Notes on Reverberation Characteristics," /. Acoust. 
Soc. ofAmer., 6 (July, 1934), No. 1, p. 51. 

3 BEKESY, G. VON: "Uber die Horsamkeit kleiner Musikraume," Annalen der 
Physik. (Series 5), 19 (1934), No. 6, p. 665. 

4 WEINBERGER, J., OLSON, H. F., AND MASSA, F.: "A Unidirectional Micro- 
phone," /. Acoust. Soc. ofAmer., 5 (Oct., 1933), No. 2, p. 139. 

5 KNUDSON, V. O. : "The Absorption in Air, Oxygen, and in Nitrogen Effects 
of Humidity and Temperature, " /. A const. Soc. ofAmer., 5 (Oct., 1933), No. 2, 
p. 112. 

6 GLOVER, C. W.: "Practical Acoustics for the Constructor," Chapman & 
Hall (London), 1933, p 165. 

7 GLOVER, C. W.: Ibid, p. 156. 

8 BAGENAL, H., AND WOOD, A. : "Planning for Good Acoustics," Methuen & Co., 
Ltd. (London), p. 270. 

9 RETTINGER, M.: "On the Theory of Sound Absorption of Porous Mate- 
rials," /. Acoust. Soc. Amer., 7 (Jan., 1935), No. 3, p. 188. 

10 HANSON, O. B.: "Planning the NBC Studios for Radio City," Proc. I. R. E., 
20 (Aug., 1932), No. 8, p 1296. 



ON THE RELATION BETWEEN THE SHAPE OF THE 

PROJECTED PICTURE, THE AREAS OF VISION, AND 

CINEMATOGRAPHIC TECHNIC* 

B. SCHLANGER** 



Summary. A shape of projected picture is suggested differing from the present one 
in respect of its conforming more nearly to the contours of the central- and peripheral- 
vision areas of the eyes. In addition to the physiological advantages of such a shape, 
greater flexibility and effectiveness of artistic composition in cinematography and in 
projection would be achieved. The present shape limits the size of the photographic 
images and details of the picture, which would be considerably enhanced in the broader 
picture shape herein described. The possibility of cinematography of a type that 
would more nearly depict what the eye sees in real life is discussed. 

Motion pictures today are not displaying the full dramatic force 
of which they are capable. Making them audible was an important 
addition to the art, but there is yet much to be done to establish the 
motion picture as the dominant instrument for recreating real life in 
the theater. The work yet to be done deals principally with condk 
tions under which the cinematographic and projection technic of the 
motion picture must be pursued. The cinematographer's "tools" 
consist of (1) a picture shape having a ratio of three equal parts in 
height to four equal parts in width; (2) a 35-mm. film width; (3) 
restricted camera angles. The projectionist's "tools" consist of 
(1) a 35-mm. film width; (2) a limited screen size; and (3) an audi- 
torium poorly adapted to the purpose of effectively presenting the 
motion picture. 

Wide-range sound reproduction and methods of acoustically 
correcting auditoriums were developed to give sound a more natural 
and realistic effect. Now new developments are needed just as 
much to make the motion picture seem more natural and realistic. 
The tools that the cinematographer and the projectionist have do not 
readily lend themselves to further development toward such an end. 
Better productions would be forthcoming as the result of an impetus 
derived from a more realistic method of motion picture portrayal. 



* Presented at the Spring, 1934, Meeting at Atlantic City, N. J. 
** New York, N. Y. 



402 



SHAPE, AREA, AND TECHNIC OF PICTURES 403 

To make the motion picture seem more natural and real, an illu- 
sion must be created to cause the spectator to feel as though he were 
within the confines of the space within which action is being por- 
trayed. Considerable research has been carried on in attempting to 
achieve a third-dimensional effect in the screen image, and thus attain 
greater realism. Such studies are still being carried on. Yet, a per- 
haps more vital consideration, that of "projecting" the viewer into 
the scene of action, a most necessary requisite to the feeling of realism, 
has received little attention. Notably, the shape of the motion 




FIG. 1. Illustrating the accurate focusing of images in the peripheral-vision 
areas, in present-day pictures. 

picture and the cinematographic arrangement therein are the im- 
portant clues toward attaining this aspect of realism. 

The name "motion picture" is somewhat misleading, and is not 
truly expressive of the science and art of cinematography. The word 
motion is self-explanatory, intimating action, but the word picture 
must be analyzed in its application to the cinema art. A picture may 
be a painting, a photograph, or a sketch to be held in the hand, hung 
upon a wall, or to be bound into a book. It is a work of art, recording 
a past incident in life. As such it has its value. Its shape and pic- 
torial composition are optional, subject perhaps only to the requisite 
of being artistic. The motion picture, on the other hand, should not 
be a physical document or record of a past event. The viewer of 
the motion picture should not be looking at a picture or through a 



404 B. SCHLANGER [J. S. M. P. E. 

picture frame; rather he would like to feel as though he existed at 
and when the action is supposed to be taking place, even though the 
"subject" may be past history. By a sheer stretch of the imagina- 
tion, it may be possible to project oneself into time and space by look- 
ing at a still picture; the element of motion in the motion picture 
makes imagining in this sense easier for the spectator, but it is not yet 
easy enough. The appeal of the motion picture will strengthen as 
this ease-to-imagine increases. 

The motion picture shape can not be thought of in the same sense 
as the shape of a vase, a tree, a building, or a work of art. A physical 
artistic form has a definitely committed shape, the outline of which 
should be pleasing to the eye by virtue of its beauty of line and pro- 
portion. As for the motion picture, it is important that it should 
assume a shape that makes the viewer least conscious of an obviously 
committed outline or shape. Therefore, the particular beauty of 
form and proportion of its shape are of no primary importance. As 
contrary as it may seem, it is the apparently "shapeless shape" that 
is most adaptable for the motion picture a shape that would make 
the viewer least conscious of a limited boundary. There are, of 
course, special instances when the motion picture can be restricted 
to a definite and committed geometrical form, as when the viewer 
is supposed to be looking through an aperture formed by a window, 
door, or other opening. But instances utilizing such an aperture 
effect are few, and act only as an occasional accent in any one 
production. 

To arrive at what would be a so-called "shapeless shape" for the 
motion picture, it is necessary to delve into the facts of physiological 
optics affecting this problem. In order to project the viewer into 
the scene of action of the story being unfolded, what he sees and 
how he sees it with his optical mechanism must be analyzed. Does he 
see in the motion picture unfolding before him what he would see 
actually existing at the scene of action? The motion picture should 
present as near a duplication as possible of what he would see by 
means of his optical mechanism in actual, real life. The same effect 
as is encompassed within the field of view by the human eyes should 
be visualized upon the screen. This field of view has an outline or 
shape, the characteristic of which should be used as a basis for the 
shape of the motion picture, but just how this shape may be ad- 
vantageously employed can be better appreciated by analyzing the 
human optical mechanism as it affects the problem. 



May, 1935] SHAPE, AREA, AND TfiCHNIC OF PICTURES 405 

Helmholtz in his Physiological Optics states, "The eye is an optical 
contrivance of remarkably wide field of view, but it is only within a 
very limited part of this field that the images are clear-cut. The 
entire field is like a drawing which is carefully executed to delineate 
the most important central part of the picture while the surroundings 
are simply sketched in more and more lightly and less distinctly out 
toward the borders. He adds that, "In spite of the vagueness of the 
broad field of view, the eye is capable of taking in a rapid glance of the 
main features of the whole surroundings and of noting immediately 
the sudden appearance of new objects in the remoter parts of the 
field." From this observation, it can be seen that the visual ap- 
preciation of our external world is accomplished by direct and indirect 
vision. The direct vision is known as distinct or central vision, the 
indirect vision as indistinct or peripheral vision. The motion pic- 
ture takes no cognizance of peripheral vision. It presents only a 
restricted area of direct central vision. 

To duplicate the effect of natural vision, the motion picture must 
undergo two changes. First, its shape must conform more nearly to 
the shape of the natural field of vision, and, second, there must ap- 
pear within this shape areas of both central and peripheral vision. 
Merely changing the shape of the motion picture alone, and not in- 
cluding within it areas of peripheral vision, will not suffice. Including 
peripheral vision will not only assist in "projecting" the viewer into 
the scene of action of the motion picture, but it will also greatly en- 
hance the art of cinematographic expression. Peripheral vision in 
real life serves as a transition for blending the sharp contrasting 
details directly in front of the eyes into the complete obscurity that 
exists behind the head of the viewer. The motion picture should 
possess such an effect; but at the present time, viewing a motion pic- 
ture can be compared to viewing our surroundings constantly as 
though we had horse-blinders attached to the sides of our heads. 
vSuch an effect would be annoying in real life, and there is good reason 
to believe that in the case of the motion picture it has an equally in- 
complete and restless effect. 

The present motion picture shape completely precludes the possi- 
bility of including peripheral-vision effects at the extreme sides, 
because every bit of its limited width is needed for action portrayal. 
As will be noted in Fig. 1, this condition always demands sharply 
defined images and details, having the characteristics of direct vision, 
out to the very edges of the picture, where indistinct vision should 



406 



B. SCHLANGER 



[J. S. M. P. E. 



exist. With a more broadly shaped screen more area could be al- 
lowed toward the extreme edges to introduce peripheral -vision effects. 
Yet, as stated before, a wider screen-shape improperly used would 
still be ineffective, as in the case of the "grandeur" screen which was 
recently given a trial use. Here the mistake was made of placing 
distinct central-vision images in sharp contrasting outlines out at the 

extreme sides. Not only did the 
images appear unreal, but they 
also caused the spectator the 
annoyance of having to shift his 
head from one extreme end to 
the other in order to follow the 
sharp, distinct images and details 
stretching across the entire screen. 
Had peripheral- vision areas been 
employed, the difficulty of the 
spectator in apprehending the en- 
tire screen at one time would 
have been reduced. 

The shape of the field of vision 
has a decided horizontal empha- 
sis, and has a proportion that 
makes the horizontal dimension 
slightly more than twice the ver- 
tical. Fig. 2 shows the outline 
of the field of vision. The ir- 
regular shape is caused by the 




FIG. 2. ( Upper} Contour of the field 
of vision of the eyes; (center} shape of 
present screen superimposed upon the 
field of vision; (lower) a suggested 
shape of picture conforming more 
nearly to the contour of the field of 
vision. 



bone structure of the head and 
by the characteristics of the 
retina. 

The shape of the motion pic- 
ture could not for practical reasons follow exactly the irregular 
outline of the field of vision as shown in Fig. 2. But its shape could 
be a rectangle, the relative horizontal and vertical dimensions of 
which could be suggested by the extreme dimensions of the field of 
vision. Fig. 2 shows at the top the outline of the natural field of 
vision, below which is shown the same field of vision with the 
present motion picture shape superimposed. Note the peripheral 
areas that do not come within the present motion picture. At 
the bottom of Fig. 2 is shown the field of vision with a suggested 



May, 1935] SHAPE, AREA, AND TECHNIC OF PICTURES 407 

motion picture shape superimposed. This rectangle would in- 
clude the complete field of vision within its shape. Note the 
shaded peripheral portion, and the still darker shaded portions sug- 
gested at the corners, where the picture composition would fade out 
to simulate the irregular outline of the field of vision. In order to be 
able to utilize correctly the entire width of the suggested shape, it 
would, of course, be necessary to investigate the possibility of suc- 
cessfully photographing the peripheral areas. This involves the 
problem of the lens and the rest of the camera mechanism. The 
camera must be able to record the peripheral-vision area as it should 
properly appear to the eyes. 

The direct- vision area on the suggested wide screen does not and 
should not always be centered upon the screen as shown at the bottom 
of Fig. 2. The direct- vision area may appear to one side, leaving a 
large peripheral area on one side only, rather than a small peripheral 
area on both sides. The peripheral area may appear concentrated 
near the top or the bottom of the screen. Such a varied use and 
positioning of the direct and peripheral areas would open up a new 
vista for effective cinematography. Both interior and exterior shots 
would profit by the advantages offered. When the "grandeur" mo- 
tion picture was tried, it was claimed that it was best adapted to 
panoramic exterior shots. As a matter of fact, the ultra-wide picture 
can be used effectively for a multiplicity of all kinds of exterior shots, 
including the panoramic type. 

In the case of interior shots, peripheral areas are important with 
only a few exceptions. These exceptions include large close-ups, 
views intentionally taken through a door or window, or views show- 
ing the interior of a small room. When these exceptions are in 
order, any optional, sharply outlined geometric form may be photo- 
graphed upon the film, leaving the remainder of the picture area at an 
even and fairly dark tone. This, in essence, becomes a changeable 
picture shape within a prescribed larger screen-form. As stated be- 
fore, such instances are comparatively few, and are most effective 
when used judiciously to heighten a particular dramatic moment. 

An example of the effect that could be created by the cinematogra- 
pher's having at his disposal the peripheral-vision areas in the motion 
picture might occur, for example, in a scene wherein new action or a 
new image is introduced at one side of the picture. The viewer 
would at first receive a vague, and then gradually a more distinct, im- 
pression as the image passed from the peripheral- to the direct-vision 



408 B. SCHLANGER [J. S. M. P. E. 

areas. With the present motion picture shape, the continuity of the 
action portrayal of such a scene is interrupted and delayed due to the 
obvious limitations of size and shape. 

Another advantage of including the peripheral areas would be to 
make the spectator feel more readily the position he holds in the scene 
of action, as well as the sense of space. For example, the extent of 
peripheral- vision area above, below, or to the side of the direct- vision 
area would vividly express whether the viewer is at a high point, low 
point, or to one side of the physical enclosure of the scene. The 
viewer would also be able to sense more readily the form and expanse 
of the surroundings of the scene of action. The director could at 
will "place" the spectator so as to create the best dramatic effect. 

Yet another advantage to be derived from the newly shaped pic- 
ture utilizing peripheral vision would be the possibility of including 
supplementary action within the field of vision, which might be 
entirely separated from the principal action in a particular scene. 
In such an instance, the supplementary action contributes to the 
principal action without being disconcerting, because such action 
would appear less distinct than the main action. This effect ob- 
viously can not be attained with the present motion picture. 

Changing the shape of the motion picture directly affects the size 
of the screen. Of course, primarily, the size of the screen is subject 
to the viewing distances to be accommodated in the theater audi- 
torium. The images and details upon the screen must not appear too 
small from the seats farthest from the screen. With the present 
average screen size, details are not sufficiently clear from the more 
remote seats; and if the more horizontal picture-shape were pro- 
jected upon the average present screen (i. e,, maintaining the present 
width but reducing the height), the clarity of details and images 
would be still further reduced. Therefore, in order to change the 
shape of the motion picture, the screen would have to be increased 
in size. More specifically, the screen would have to be slightly 
greater in height and considerably greater in the horizontal dimension 
to offer sufficient area for both distinct-vision and peripheral-vision 
details of sufficient size. Enlarging the screen naturally involves 
increasing the width of the film to assure clear vision of the image from 
the seats nearest the screen. 

Another factor that limits the realism of the motion picture lies in 
the relation of the screen to the auditorium. The screen always ap- 
pears small against the adjacent walls of the auditorium. The walls 



May, 1935] SHAPE, AREA, AND TECHNIC OF PICTURES 409 

or forepart of the auditorium compete strongly with the screen, es- 
pecially because such strong contrast exists between the lighted 
screen surface and the darker surroundings of the screen. In place 
of the dead black masking now used to frame the screen, a supple- 
mentary border should be used, having a shape conforming to the 
natural vision contour, which could be lighted to an intensity that 
would blend with the lighting of the auditorium and the screen. 
This screen border lighting would then serve as a transitional blend- 
ing between the walls of the auditorium and the illuminated screen 
surface. 



MECHANICAL RECORDING ON FILM* 
A. F. CHORINE** 

Summary. A method is described of mechanically recording sound upon used 
film by means of a diamond cutting stylus. The same instrument can be used for 
reproducing the sound so recorded, merely by substituting a needle for the stylus, and 
making appropriate changes in the input and output of the amplifier. The quality 
of the recordings is surprisingly good, and is proving satisfactory for certain purposes. 

The Soviet sound motion picture industry is operating under 
conditions that often make it necessary to make repeated recordings 
and to accumulate much recorded material from which to select the 
best records for the composition of the picture. When field expedi- 
tions are made, it is more than ever necessary to make many and 
varied recordings. 

There are two reasons for making a great number of recordings, 
both variants and repetitions : One is the abundance and originality 
of sound material, suitable for travelogues, in such regions as Siberia 
and Middle Asia, and the other is to assure successful reproductions, 
inasmuch as in such places there are no facilities for developing the 
films, reproducing them on the spot, and thereby ascertaining their 
quality. 

Under such conditions the material has to be developed and tested 
after the return of the expedition, and therefore the use of ordinary 
films and optical methods of recording becomes exceedingly expen- 
sive. All this has suggested the idea of inventing an instrument for 
recording sounds mechanically on a used film. The principle of 
this contrivance is very simple. Figs. 1 and 2 show the general ap- 
pearance of the apparatus. In it may be seen the mechanism for 
moving the film along the plate 1 together with a mechanical filter to 
assure a uniform motion of the film. Above the plate there is a 
recording arm 2, in which a stylus made of hard stone or diamond 
is inserted. 

* Received Aug. 20, 1934. 

11 * Director, Central Laboratory, All-Union Electrical Trust, Leningrad, 
U.S. S. R. 
410 



MECHANICAL RECORDING ON FILM 



411 



With the aid of a special contrivance 3, it is possible to adjust 
precisely the position of the sound record upon the film. Over the 
entire width of the film can be placed nearly 50 such records side by 
side. This mechanical recorder has a special dial indicating the 
ordinal number of the tracing and the micrometer for adjusting 
the stylus. For rapid transition from one groove to another during 
the sound reproduction there is a second dial by means of which the 
pick-up with inserted needle can be moved rapidly across the film. 
The same arm can be employed both for recording and for reproduc- 




FIG. 1. Mechanical recorder and reproducer. 

tion. In order to pass from recording to reproduction it is 
sufficient to replace the stylus by a needle and to inter- 
change the input and the output connections of the amplifier, thereby 
making it immediately possible to hear the recorded material and 
judge its value. For the purpose of observing the sound record 
during the process of cutting there is a small magnifying attach- 
ment, with the aid of which it is easy to look after the adjustment 
of either the stylus or the needle during their performance. Ex- 
perience has shown that a film is excellent material upon which to cut 
sound records, and is very durable in the sense of permitting a great 
number of reproductions. Since 50 tracings can be cut on the same 
film, it is possible to record many hours of material upon a reel con- 



412 A. F. CHORINE [j. s. M. p. E. 

taining only 300 meters of film. This is many times in excess of 
what could be recorded optically. In this connection, it is worth 
noting that already exposed films or films rejected for optical im- 
perfections can be utilized for mechanical recording. 

Under the conditions existing in the Soviet Union, the mechanical 
system of sound recording directly upon a film is of special interest 
not only in its application to recording instruments but also for its 
potentialities in the establishment of provincial and "kolkhoz" 
talking motion picture shows. In the U. S. S. R. there are, at present , 




FIG. 2. Close-up view of mechanism. 

nearly 40,000 travelling silent motion picture shows; in the near 
future their number will be greatly increased. The necessity for their 
adoption is urgent. Moreover, there exists a tremendous supply of 
silent films intended for rent in these travelling, motion picture shows. 
If the mechanical system of sound recording is utilized, the film with 
the sound records is run synchronously with the silent motion picture 
film, and the latter is immediately converted into a talking picture. 

In small motion picture shows this mechanical system of sound 
reproduction offers a number of practical advantages over the photo- 
electric system because of its greater simplicity. Because of the 
greater efficiency of the pick-up by comparison with the photoelectric 



May, 1935] MECHANICAL RECORDING ON FlLM 413 

cell, the necessary amplification will be reduced, and the optical 
system, especially the light source, is excluded altogether. On ac- 
count of the enormous area of the Soviet Union and because of the 
frequent lack of electric power supply, such a system of sound record- 
ing, even if used solely for the purpose of musical accompaniment, is 
already very promising. This is especially true as only a 6-volt 
battery is required, which can always be borrowed from a car or a 
tractor, both of which are nowadays easily found in any settled 
region. 

The only drawback of the system is its equality. The present engi- 
neering tendency in talking moving pictures is to assure a transmis- 
sion band up to 7000-8000 cycles, and it appears that the mechanical 
system is not a step forward in so far as quality is concerned, since the 
frequency band will actually be narrower. However, the quality of 
the system is surprisingly good, and is proving satisfactory in an 
emergency. 



THE EDUCATIONAL MOTION PICTURE OF YESTER- 
DAY, TODAY, AND TOMORROW* 

H. A. GRAY** 

Summary. The evolution of mechanical aids to learning is traced briefly, and the 
advantages of the motion picture in overcoming limitations to learning are discussed. 
Experiments for testing the efficiency of silent and sound pictures as media of instruc- 
tion are cited. 

Reference is made to some of the problems that must be solved before educational 
motion pictures can be used more widely informal public instruction. Suggestions 
are made for measuring objectively the effects of sound motion picture elements upon 
the observer by means of mechanical recordings and detailed study of the individual. 

In this paper is presented, not the point of view of an engineer, but 
that of a research worker who is concerned with the development of 
the sound picture for instructional purposes. The evolution and 
present status of the educational motion picture will be described 
briefly, and the movement that its developments have created will 
be alluded to. Also, some of the problems involved in its utilization 
will be mentioned, and an area of research calculated to enhance its 
power to stimulate human beings will be outlined. The latter point 
is of particular significance because in the future the sound picture en- 
gineer and production specialist may be envisaged as devoting con- 
siderably more attention to the problem than has been the case in 
the past. 

Since the beginnings of society the acquisition and dissemination 
of knowledge have been influenced by the facilities available for such 
activities. As a savage, man's learning was restricted to that ob- 
tained from signs, demonstrations, and experiences with the elements 
going to make up his environment. Then came sounds and language, ; 
which for a time were destined to become the chief means of propa- 1 
gating information. With the invention of writing, man was able to 
record his learnings for current and future reference ; first on clay j 
tablets, and later on the papyrus roll, and for hundreds of years, 



* Presented at the Fall, 1934, Meeting at New York, N. Y. 
* * Erpi Picture Consultants, Inc., New York, N. Y. 



414 



THE EDUCATIONAL MOTION PICTURE 415 

spoken and written language, demonstrations, and experiences were 
the modal methods of acquiring knowledge. The invention of the 
printing press gave impetus to the widening of intellectual horizons, 
and for another long period the printed page dominated, to the ex- 
clusion of almost everything else, the quality and quantity of learn- 
ing. As time went on, however, a need was felt for other devices to 
serve as learning aids. This need was gradually met by science as 
invention created a number of instruments of instruction. Labora- 
tory apparatus, still pictures, models, exhibits, charts, graphs, maps, 
stereographs, slides, silent motion pictures, phonographs, radios, and 
sound motion pictures are among the more important ones, and all 
have contributed materially to the advancement of learning. Those 
whose use depended upon the sense organ of sight became known as 
visual aids, and the phonograph and radio have acquired the title of 
auditory aids. Recently the sound motion picture has been referred 
to as a visual-auditory aid. 1 

The term visual education may be taken literally to mean the sum 
total of all visual experiences, but for practical purposes it is usually 
limited to the educational use of materials specifically created to en- 
hance the impression that an observer may acquire of a situation. 
The indications are that each device may occupy or carry on a par- 
ticular function in the classroom, that it is possible to achieve results 
with one device that can not be as satisfactorily obtained with other 
devices. The question was first of all a matter of opinion. Then it 
became a matter for scientific experimentation to decide. The 
effectiveness of still pictures, models, exhibits, charts, maps, graphs, 
and stereopticon slides has been demonstrated by their widespread 
use in all forms of educational activities. Teachers have come to 
depend upon these devices for assisting them in developing concepts 
otherwise difficult to acquire. Early experimentation, however, 
such as that conducted by Weber, 2 has shown that the effectiveness 
of materials of visual instruction depends to a great extent upon the 
degree to which they are integrated with verbal concepts and printed 
supplementary materials. 

About the beginning of the present century, the late Thomas A. 
Edison 3 predicted that the silent motion picture would in time oc- 
cupy a prominent place in education. It was not until some years 
later, however, that educators all over the country became enthusi- 
astic over the educational potentialities of the film. Its use led to 
considerable experimentation regarding its instructional effective- 



416 H. A. GRAY [j. s. M. P. E 

ness. Wood and Freeman 4 found in their controlled experiment 
with 11,000 school children, that the group instructed through the 
use of motion pictures achieved about 17 per cent more in mean gain 
in geography tests and about 11 per cent more mean gain in general 
science tests than the members of their control groups. Knowlton 
and Til ton 6 reported gains of 19 and 12 per cent, respectively, in 
favor of those experimental groups that had seen The Chronicles of 
America before taking tests in American history. These early ex- 
periments were substantiated by similar ones conducted in England, 
where Burt, Spearman, and Philpot 6 concluded from their investiga- 
tion that the motion picture should be an integral part of the educa- 
tive process. Some time later, Freeman 7 and his collaborators in 
America described the motion picture as having a distinct educational 
value in the subjects of nature study, geography, handwork, high- 
school science, home economics, English, health, and even handwrit- 
ing, their conclusions being based upon the experiments that they 
conducted as a further check upon the film's efficacy. 

With the advent of the sound motion picture, additional experi- 
mentation was undertaken. A testing project, supervised in part 
by the U. S. Office of Education, indicated that the sound film was 
about twice as rich in instructional values as its predecessor, the 
silent film. 8 About the same time an independent investigation, con- 
ducted at Columbia University with adult graduate students as sub- 
jects, showed a twenty-minute sound picture to be a significantly 
more effective stimulus than longer periods of time spent on discus- 
sions, writings, and lectures. 9 A third experiment, conducted in 
England under the auspices of the Middlesex School Committee, in- 
dicated not only substantial learning increments on the part of the 
pupils but definite interest and enthusiasm from the teachers partici- 
pating. 10 The Arnspiger experiment 11 carried on in the five cities of 
Schenectady, New York, Elizabeth, Camden, and Baltimore, and 
involving sixty-four schools with some 2200 pupils, showed that the 
groups using the pictures achieved 25.9 per cent more in natural 
science and 26.9 per cent more in music. In addition, the sound 
picture groups retained more of the knowledge thus gained over a 
period of three months. Other testing projects conducted under the 
auspices of Columbia, Harvard, and New York Universities have 
substantiated for the most part the previous findings involving the 
use of the sound film. 12 ' 13 ' 14 

The recent investigation, financed by the Payne Foundation to 



May, 1935] THE EDUCATIONAL MOTION PICTURE 417 

study the effect of theatrical motion pictures upon children likewise 
indicated the effectiveness of the medium for shaping attitudes, 
stirring up emotions, molding morals, and generally influencing be- 
havior. 15 The sound motion picture has been described as being one 
of the most influential forces in contemporary social life; a fact easily 
determined by study of the success of advertising, propaganda, and 
other types of films designed to mold public opinion. 3 

Many teachers colleges and normal schools are now requiring 
their students to develop skill in the use of visual aids, including 
motion pictures, as an essential part of their training. Other teacher- 
training institutions offer courses in such instruction as electives. 
At the present time, thirty-one colleges and universities list such in- 
struction as separate courses, and twenty-one university extension 
divisions maintain visual instruction service chiefly with motion 
pictures. 16 

During the past few years a great number of books and articles 
in educational periodicals dealing with aspects of education by mo- 
tion pictures have appeared. The magnitude of this activity is 
attested by a recently published bibliography listing fifty-one books, 
thirteen booklets, thirteen M.A. and Ph.D. theses, and hundreds of 
articles, selected from an even more extensive list. 17 

The movement has grown to the point where the state and na- 
tional governments of practically all civilized countries are taking an 
active part in the production and distribution of educational motion 
pictures. 3 In America, the Federal government is probably the 
largest single sponsor of educational motion pictures. The Bureau 
of Mines, the Department of Agriculture, and other governmental 
departments have prepared an enormous amount of picture material 
for imparting information upon matters with which they are con- 
cerned. Without doubt, the government's activity in this field has 
been of great value to many industries and individuals throughout 
the world. 

The preceding indicates that educators internationally have recog- 
nized the power of the celluloid film, particularly the sound picture, 
to overcome physical and human limitations to learning physical 
with respect to the restrictions of time and space, and human with 
regard to the inability of the human senses to perceive objects and 
relations too abstract in nature or too finite in dimension for human 
conception. 

The sound motion picture is one of the most powerful motivating 



418 H. A. GRAY [J. S. M. P. E. 

devices in existence for stimulating such learning and its constituent 
attitudes and appreciations; and if created and utilized intelligently 
it can be made to serve the purposes of education as no other device 
available. With it entire new vistas are opened up to the learner's 
perception. The streaming of protoplasm through cell tissues, ani- 
mated molecular action, the sight and sound of a human heartbeat, 
the striking colors and sounds of under-sea flora, slow motion pictures 
of bird's wings in flight, ultra-rapid pictures of growing plants, x-ray 
portrayals of mineral deposits, and many other natural phenomena 
may be seen and heard by means of the various sound-photographic 
technics that have been developed. 

Then there are the more subtle elements of social and economic re- 
lations, which can be vividly portrayed in life situations with a chal- 
lenging appeal to the learner's attention and interest. The social 
studies, particularly, have become beneficiaries with these materials. 

Schools not possessing elaborate plant and equipment facilities for 
comprehensive study activities, need be no longer so handicapped. 
With a comparatively small capital outlay, the sound film can bring 
to the individual classroom unlimited study materials. The mes- 
sages of world leaders, difficult and expensive laboratory experiments, 
exhibits of museum and art collections, travel experiences in fact, 
almost every conceivable element of a world environment now may 
be made available for boys and girls with previously restricted edu- 
cational opportunities. 

So much, in brief, for the development and present status of the 
educational motion picture. Its advantages will now be objectified 
by presenting a film devoted to the explication of sound phenomena. 
Note how the medium is employed to portray the mechanism of 
hearing; how such intangibles as sound velocity, refraction, range, 
intensity, attenuation, low and high frequencies, reverberations, and 
focusing are treated by the marvels of animation and sound effects. 
Compare this method of presentation with the effectiveness of con- 
ventional methods in the physics classroom, and draw your own con- 
clusions. 

(At this point Dr. Gray presented the sound picture "Fundamentals of Acoustics," 
courtesy of Erpi Picture Consultants, Inc.) 

We shall now refer to some of the major problems that must be 
solved before the sound film can render the optimal educational ser- 
vice of which it is potentially capable. Some of these problems are 



May, 1935] THE EDUCATIONAL MOTION PICTURE 419 

common both to the producers of educational pictures and to the 
manufacturers of sound projection equipment. Other problems are 
the chief concern of either the picture producer or of the equipment 
manufacturer. Still others rest squarely upon the educational au- 
thorities themselves. 

Part of the failure of educators to use the silent picture more ex- 
tensively was due to the frequent failure of the films offered to fit 
into any particular course of study. Also, such pictures often dupli- 
cated what the teacher could do quite as effectively with other de- 
vices. Many films lacked either logical or psychological continuity, 
or both. Few were arranged with definite teaching objectives in 
mind, and still fewer were accompanied by printed supplementary 
material for the teacher's edification, not only on the picture's pro- 
jection, but on its integration with other study activities as well. 
This is primarily the producer's problem, and Erpi Picture Consul- 
tants have attempted to solve it by developing objective standards 
for the production of instructional sound films ; surveying the various 
fields of educational effort to determine wherein the sound film could 
make contributions; analyzing courses of study for the selection of 
picture topics ; preparing scenarios for the topics ; meeting the stand- 
ards and criteria set up ; retaining production specialists for the prepa- 
ration of pictures; testing the classroom effectiveness of the sound 
films produced; cooperating with school authorities in developing 
utilization programs; and arranging supplementary printed material 
to accompany each picture. 

A second major problem concerns the manufacturers of sound 
equipment. Ultimately auditorium equipment should be in every 
school having such an assembly space. This will serve for the in- 
struction of large groups of students and occasionally for entertain- 
ment purposes. However, the big future of the sound picture will 
be in the individual classroom, where the teacher may on a moment's 
notice project upon the screen a vitalized part of his lesson plans. 
The specifications for such equipment obviously call for low initial 
cost and upkeep, portability, simplicity of operation, rigidity of con- 
struction for reliable performance, and, of course, 16-mm. sound-on- 
film reproduction of good quality. 

These two problems must be solved simultaneously, for, until ade- 
quate pictures and equipment are available, producers can not sell 
prints nor can the manufacturer find an outlet for his projectors, 
however good either may be. 



420 H. A. GRAY [j. s. M. P. E. 

A third problem concerns the responsibility of educational authori- 
ties to plan for the extension of instruction by sound pictures. This 
involves first of all the training of teachers in the use of projection 
equipment and the handling of sound film. Each teacher-training 
institution should include such instruction in a course to be called 
perhaps, "Special Methods with Mechanical Aids." Teachers al- 
ready in service may likewise receive instruction in the utilization of 
such devices. State departments of education can well promote the 
organization of local departments of audio- visual instruction, and 
sponsor the initial purchase of equipment and films for communities 
unable to bear the financial burden entailed. An inference regarding 
the state's responsibility for a comprehensive visual education pro- 
gram is made by Dr. C. M. Koon, senior specialist in radio and mo- 
tion pictures of the United States Office of Education, who recently 
remarked : 

"... Probably one of the most important problems facing workers in visual educa- 
tion is to determine means whereby the state can be made aware of the vast re- 
sponsibility resting upon it to make provision for the use of films in all schools 
within its domain. As soon as visual instruction is sponsored by the state, we can 
rest assured that we shall have a much better educated populace." 17 

Another problem common to all concerned with the production and 
use of the educational sound picture is that of research devoted to 
discovering how, when, and why elements of sight and sound affect 
the individual visibly or otherwise, enough to modify his thinking 
or behavior. The sound picture with is human, animal, elemental, 
and mechanical sound stimuli; trick, natural, and animated pho- 
tography of form stimuli; and its black, white, and other color stim- 
uli ; effect in the individual glandular, muscular, and neural systems ; 
changes known as interests, attitudes, learnings, emotions, and ap- 
preciations. Past attempts to measure these characteristics of men- 
tal development fall into three general classifications. The first is 
individual observation, in which the experimental examinee is sub- 
jected to direct interrogation, participates in interviews, engages in 
discussions, and has his conversation analyzed. 

A second type of experimentation has utilized paper-and-pencil 
tests and procedures. The individual's reaction to word association 
checklists, reading choices, questionnaires, opportunity for self-analy- 
sis, subjective questions, personal preferences, "yes-no" multiple- 
choice problems, contrasting statements, superstitions, and speed 



May, 1935] THE EDUCATIONAL MOTION PICTURE 421 

tests have been reported in voluminous detail along with records of 
his factual and vocabulary knowledge. 

The third classification describes the technics that have been util- 
ized in mechanically recording the subject's respiratory and pulse 
rates, his blood pressure, muscular tension, vocal responses, psycho- 
galvanic reflexes, facial expression, and resistance to interference 
while subjected to selected stimuli. 

Because of the experimental limitations, these approaches have 
been foredoomed to partial if not complete failure in attempting to 
solve the problems with which they were concerned. Until such 
procedures are developed in combination, together with a detailed 
study of the individual and of the stimulus to which he reacts, the 
recording of such degrees of failure is likely to continue. 

At the present time, numerous tests and methods of gaining in- 
formation about the individual are available. Each of these has 
some merit, but when taken singly and without respect to the in- 
dividual as a whole their value is lessened decidedly. However, 
some knowledge of a subject's reputation, imagination, attitudes, 
morality, interests, inhibitions, home life, general character, aggres- 
siveness, sexual traits, confidences, emotions, appreciations, com- 
munity life, esthetic reactions, speed of response, sociability, physi- 
ology, and suggestability, along with some measure or index of his 
mechanical, abstract, auditory, and social intelligence, will provide 
valuable clues and reasons for his behavior in the experimental situa- 
tion. Nevertheless, most of such information would be highly sub- 
jective and difficult to correlate, and the author feels that some mea- 
sure of the individual's abstract and auditory intelligence, his reac- 
tion speed, and his electrical resistance and capacity would suffice for 
beginning the experimentation proposed. This information should 
be obtained before the individual reacts to the stimuli of the sound 
picture. 

It is further believed that the recording of the individual's reac- 
tions to the sound film in synchronization with its presentation, and 
consisting of his respiration and pulse rates, muscular tension, psycho- 
galvanic reflexes, facial expression, and brain oscillations would pro- 
vide a much more valid and reliable measure of his whole response 
than anything yet attempted. The meaning of the variations and 
correlations obtained by such recordings would be clarified by paper- 
and-pencil procedures, discussion, direct questioning of his thoughts 
and feelings as to elements in his visual-auditory experience, and 



422 H. A. GRAY [J. S. M. P. E. 

the data that were obtained prior to the presentation of the 
picture. 

Similarly, an analysis should be made of the sound picture stimuli 
before they act upon the subject. A detailed description of the color 
elements, their location, contrasts, and combinations, should be pro- 
vided. An enumeration and description of the sound elements, in- 
cluding their kind and source, also would be necessary, together with a 
classification and description of the form elements, their appearance, 
location, and mode of presentation in each scene. A synchronized 
record of the volume of sounds and the intensity of the light reflected 
from the screen, would provide an accurate index of the strength of 
the composite sound and light stimuli involved, and would be of 
value in interpreting the many cause-and-effect relations that un- 
doubtedly would be discovered when all the data were grouped into 
comparable units for detailed analysis. 

(Dr. Gray terminated his presentation by projecting the picture "Sound Waves 
and Their Sources," courtesy of Erpi Picture Consultants, Inc.) 

REFERENCES 

1 DEVEREAUX, F. L. (et al.): "The Educational Talking Picture," Univ. of 
Chicago Press (Chicago, 111.), 1933. 

2 WEBER, J. J.: "Comparative Effectiveness of Visual Aids," Educat. Screen, 
Sec. Ill, 1926. 

3 GRAY, H. A.: "Can Educators Profit from Industry's Experience with the 
Motion Picture?" Educat. Screen, XII (Apr., 1933), No. 4, p. 101; XII (May, 
1933), No. 5,p.l23 

4 WOOD, B. D., AND FREEMAN, F. N.: "Motion Pictures in the Classroom," 
Houghton Miffiin Co. (New York, N. Y.), 1929, p. 215. 

6 KNOWLTON, D. C., AND TILTON, J. W.: "Motion Pictures in History Teach- 
ing," Yale Univ. Press, (New Haven, Conn.), 1929. 

6 MARCHANT, J. (ed.): "The Cinema in Education," George, Allen, and Erwin, 
Ltd. (London), 1925. 

7 FREEMAN, F. N. (ed.): "Visual Education," Univ. of Chicago Press (Chicago, 
111.), 1924. 

8 "Sound Pictures as a Factor in Education," Fox Film Corp. (New York, 
N. Y.), 1931. 

9 EADS, L. K., AND STOVER, E. M.: "Talking Pictures in Teacher Training" 
Erpi Picture Consultants, Inc. (New York, N. Y.), 1931. 

10 WALTON, H. M. (ed.): "Sound Films in Schools," The Schoolmaster Pub. Co. 
(London), 1931. 

11 ARNSPIGER, V. C.: "Measuring the Effectiveness of Sound Pictures as 
Teaching Aids," Teachers College, Columbia Univ., Bur. of Pub. (New York, N. Y.), 
1933. 

12 WESTFALL, L. H.: "Study of Verbal Accompaniments to Educational Mo- 



May, 1935] THE EDUCATIONAL MOTION PICTURE 423 

tion Pictures," Teachers College, Columbia Univ., Bur. of Pub. (New York, N. Y.), 
1934. 

13 RULON, P. J.: "Sound Motion Pictures in Science Teaching," Harvard 
Univ. Press (Cambridge, Mass.), 1933. 

14 CLARK, C. C.: "Talking Movie and Students' Interests," Science Educa- 
tion, XVII (Feb., 1933), No. 2, p. 17. 

15 FORMAN, H. J.: "Our Movie Made Children," Macmillan (New York, 
N. Y.), 1933. 

16 "Visual Instruction Directory," Dept. of Visual Instruction, Nat. Ed. Assoc. 
(Laurence, Kansas), 1933. 

17 KOON, C. M. (et a/.) : "Motion Pictures in Education in the United States," 
Univ. of Chicago Press (Chicago, 111.), 1934. 



THE THEATERGOER'S REACTION TO THE AUDIBLE 
PICTURE AS IT WAS AND NOW* 

MORDAUNT HALL** 

Summary. A brief description, in narrative fashion, of the improvements thai 
have come to the motion picture since the day of the silent, with particular reference to 
dramatic tastes and preferences , from the point of view of the critic. 

Not only has the linking of sound with motion pictures been re- 
sponsible for the employing of infinitely more intelligent players than 
one usually found in silent films, but it has taught audiences to a 
great extent to appreciate a higher order of screen entertainment. 
The talking picture is now a diversion that appeals to all manner of 
men and women. Yet, at the outset, only a few years back, when 
Warner Brothers released their first Vitaphone features, sound was 
thought by several of the more astute and moneyed motion picture 
producers to be but a passing fancy on the part of the public. Many 
a writer scoffed at the idea of coupling the microphone and the camera 
and even that celebrated dramatist, Pirandello, took a fling at the 
notion of having speech come from a flat surface. Others contended 
that it sounded the death knell of the cinematic art. They bemoaned 
the new invention, or relatively new because of its sudden popularity. 
Undoubtedly the first sound pictures offered much to be critized ad- 
versely, and as the screen critic of the New York Times, I was fre- 
quently asked whether these squawking affairs would continue. 
Some declared that they meant the end of the great industry. They 
called attention to the ludicrous lisping of the actors, and the theater 
managers did not help matters by insisting upon having the speech 
come forth in deafening tones. They boasted that every seat was a 
front-row seat, and saw to it that those in the rear of their houses 
were never in doubt as to what was being said by the characters. 

Even though the voice was the thing later in Lawrence Tibbett's 
film The Rogue Song, this baritone's vocal efforts were reproduced 



* Presented at the Fall, 1934, Meeting at New York, N. Y. 
** New York, N.Y. 



424 



THE AUDIBLE PICTURE 425 

with such fierce loudness that one felt that the theater management 
hoped to have the premiere of the feature heard not only outside the 
theater in Times Square, but across the Atlantic. This was somewhat 
distressing to writers like myself, who favored the audibility of the 
screen, but soon it was evident that as each new dialog film was pre- 
sented a steady improvement was revealed. True, there were the 
shortcomings of the players, such as those pretty blondes who should 
have been seen and not heard, for their baby voices or their twangy 
uncultured intonation was not a little irritating. Many of these at- 
tractive feminine players eventually found themselves in the discard, 
their places being taken by experienced stage performers. 

Then, too, there was in the beginning of the sound offerings the 
disconcertingly poor synchronization. Frequently the voices ap- 
peared to be coming from the feet of the players, and this gave the 
opponents of sound further chances to heap ridicule upon the talking 
pictures, as did also the frightening explosive outbursts that were 
emitted from the screen all too frequently. Most of the players in 
those times seemed to be declaiming, speaking their lines as rapidly 
as possible, either because they feared that they would not be able 
to get rid of their speeches in the given time, or because they were in 
terror of forgetting what they had to say. Directors who had been 
in the habit of shouting through a megaphone were forced to do 
everything, by signals while the sound appliance was working, and 
many a dainty graduate from a manicure shop or a lunch-counter 
failed to grasp what the director wanted her to do, with the result 
that the director either ordered a retake, or tore his hair as silently 
as possible. During this state of affairs it was not astonishing that 
intelligent persons did more jeering at the dialog films than they had 
done at silent ones. There were those patrons of the films who had 
found silence a boon and a blessing. If the feature was poor enter- 
tainment they at least could take a nap. The noise of audible pic- 
tures kept these people awake. 

Those Russians who had done so well with mute works were by no 
means enthusiastic over giving voices to the shadows, but as years 
passed the Soviet film makers discovered that a resourceful and an 
imaginative director could inculcate art into a sound film as well as 
he could in a silent subject. In fact, he had a double instrument with 
which to appeal to his audiences. 

Now and again in the days of five or six years ago, features ap- 
peared with periodical outbursts of speech, followed by silent episodes. 



426 MORDAUNT HALL [J. S. M. P. E. 

These films were really awful. One or two producers later took a 
chance at a silent film, but except in the case of Chaplin's City Lights, 
silence was no longer golden. Even the artistic Chaplin, who cari- 
catured dialog films in his last comedy, took full advantage of certain 
sound effects. Chaplin was wise enough to know that his wordless 
film would sell all over the world, and it is hardly likely that dialog 
in his subjects would be as successful as pantomime, especially Chap- 
lin's own work. Few of his hosts of admirers would ever want him to 
speak in his comedies. His films, however, are the exception that 
proves the rule. 

So far as other more recent silent productions are concerned, they 
seem, after one has become accustomed to hearing speech from the 
screen, to be, if anything, more absurd than any of the first vocalized 
films. In the old product title writers committed many sins. They 
were fond of repeating pet phrases. How often one would read in 
the text, ''Came the dawn," "Just around the corner dwelt," "The 
chatelaine of the manor was," and "It's a far cry from" ! Even that 
was all very well until the sound productions showed progress; but 
viewing a mute offering then often resulted in great disappointment, 
as one had by that time become accustomed to hearing the characters 
speak and, therefore, refrain from gesticulations that had become 
quite common in even the best of silent films. It has been my ex- 
perience invariably to find that the old gems of the past were far in- 
ferior to the present-day worthy talking picture. In fact, it often 
happened that when one saw the silent features over again they 
seemed quite inferior. Many believe that, in order to achieve satis- 
factory cinematic values, it is better to have a minimum amount of 
dialog. This idea, in my opinion, applies only to isolated instances, 
for in the majority of cases the real vitality of a film is now the dialog. 

A remarkable subject that offered one of the most impressive com- 
binations of the microphone and the camera was The Invisible Man. 
As one now listens to the extraordinary natural quality of the sound 
one can not but marvel at it, especially when he harks back only a 
year or so and remembers the curious things that happened in the 
studios. Monta Bell, the director, told me that in experimenting with 
sound films he had made the amazing discovery that a lump of sugar 
dropped into a tea-cup came forth from the screen like the sound of a 
fifteen-inch gun. 

Talking pictures rival politics as after-dinner chatter. It is strange 
that one has only to sit down a few minutes before he hears the 



May, 1935] THE AUDIBLE PICTURE 427 

conversation drift to talking pictures, and once there it usually en- 
dures. 

When I read of somebody damning a picture because it is nothing 
but a photograph of a dramatic work, I feel that there is something 
to say for the fact that we are now able to get even a photograph of a 
play, for that is surely preferable to some of the films that find their 
way out of the studios. But the poor stuff that is produced is not 
always quite the fault of film makers, for unfortunately, even in this 
year of grace, there are still many patrons who dote upon happy 
endings and dislike intensely to have their screen favorites pass away 
at the end of the story. One also finds that apple blossoms and love 
still go together with amazing frequency, and because of the desire 
to please unthinking audiences there exists no suspense in film tales. 
These are some of the heritages of the past. And strangely enough, 
educated persons with alert minds have frequently revealed to me 
their poor taste in pictorial narratives. But, as I have already re- 
marked, motion pictures have improved immensely and it is to be 
hoped that the silly effusions will become fewer and fewer. 

When the sound in pictures became what it is, or has been for the 
last year or two, some of the producers revealed a penchant for plant- 
ing episodes in wash-rooms, and on one occasion a distinctly vulgar 
use was made of the marvelous invention. This peculiar type of 
humor was undoubtedly inspired by something that happened in a 
play. Incidentally, so far as coarseness is concerned, one might say 
that the stage is way ahead of motion pictures. True, there are 
wearying instances of stuff made to appeal to the clodhopper, with 
vulgar outbursts that are both reprehensible and silly. 

If the producers, instead of concentrating so often upon suspenseless 
twaddle, were to turn their attention to something worth while, it 
would be doing something comparable to the work done by the inven- 
tors. As the uppermost thought of most picture makers when se- 
lecting a story is to turn out something that will bring home the 
bacon, there ought to be a few philanthropists who might encourage 
the production of Shakespeare's works, and other classics. Also, 
many persons over here and still more in Britain, Australia, Canada, 
South Africa, and so forth, would delight in hearing and seeing some 
of the Gilbert and Sullivan works on the screen. Such features might 
not bring in returns equal to those of Forty-Second Street or Chained, 
but they might, if really well done, surprise their producers. As to 
Shakespeare, granting that the audiences might not be numerous 



42S MORDAUNT HALL [J. S. M. P. E. 

enough to pay for showing films of the bard's- plays in the big cinemas, 
such productions might easily be screened in schools and colleges. 

I do not recall whether when I addressed this gathering several 
years ago, I commented upon the number of performances given in 
pictures; it is a marvelous contrast to the stage. When Grumpy, 
with the English player Cyril Maude, was put on at the Paramount 
here, Mr. Maude was very proud of having given 1300 performances 
of the play in various sections of the world. Had he acted continu- 
ously, which he did not, I believe that I am right in saying that the 
1300 performances would have taken more than three years. Now, 
take the picture of the play suddenly bursting forth in various centers 
in huge theaters, often for five and six and seven shows a day. It 
has been computed that before Mr. Maude's shadow got through 
playing Grumpy it had given something like 150,000 performances. 

There was a time when it was estimated that the weekly atten- 
dance in the United States at motion picture houses was 115,000,000. 
This figure has not been touched since the prosperous times. 

Spectators have now reached a point when they expect perfection 
in sound recording, so much so that they would complain of the 
slightest technical deficiency. Although Eugene O'Neill's Strange In- 
terlude was cut down to suit the commercial length of pictures, and 
therefore much of the brilliant play was left out of the film, it is in- 
teresting to note that the idea of "asides" being spoken was far better 
suited to the screen than to the stage. On the stage the players were 
called upon to speak these asides, and it was not entirely successful 
because their between-the-teeth utterances were not always clear. 
For the screen these asides were recorded before the players were 
photographed for the scene, and in the picture their lips were still. 
It was, I believe, in Blackmail, an English picture, that the first 
thoughts were heard from a character, and this brings to mind the 
fact that the producers of the film of Dreiser's American Tragedy 
would have added greatly to their picture had they made audible the 
thoughts of the young murderer while he was in the telephone booth 
talking to his victim and asking her to meet him. 

Sound or audibility was employed in a new fashion in Jesse L. 
Lasky's Power and the Glory, the treatment of which was termed 
"narratage." The story began at the end; the incidents were not 
unfurled in the usual flash-back way, but told by living characters 
about a husband and wife who were both dead. The opening scenes 
revealed the funeral of the central character and also what certain 



May, 1935] THE AUDIBLE PICTURE 429 

people thought of him. After that the story came from several 
angles, showing what led to the suicide of the principal character and, 
prior to that, of the suicide of his wife. 

Notwithstanding the superior taste of audiences in motion pictures 
since talking films were made, it is regrettable that narratives en- 
dowed with much subtlety are seldom successful and also that satiri- 
cal ventures invariably fail to bring in much to the box-offices. 
Hokum is popular, but, I am glad to say, it is an improved type of 
hokum over what was usually offered in the silent effusions. Hence, 
even in this respect audibility has accomplished something. 

Ernst Lubitsch, that masterful German director, was, if memory 
serves correctly, the first to blend the sound of machinery with 
music, which made a curiously effective rhythm. He placed Jeanette 
MacDonald in a train coach, and had her sing to the accompaniment 
of a combination of music and the sounds of the speeding train, which 
included the whistle warning, the roar of the coaches and the jogging 
sound of the wheels. And while this director produced some fine 
examples of silent pictures, there is no doubt but that sound gave 
him something more with which to conjure. As somewhat of a con- 
trast to Lawrence Tibbett's Rogue Song, there was recently the opera 
singer, Grace Moore, in a picture called One Night of Love, which 
even at a time when the public and the press had become accustomed 
to very high quality of reproduction, aroused great praise for the fine 
recording, particularly of the singing. Little did audiences expect 
such excellent results when they beheld that first musical film, Broad- 
way Melody, wherein there was the absurd idea of having theatrical 
troupers singing in a hotel room accompanied by Paul Whiteman's un- 
seen orchestra. Think of that outstanding picture, Cavalcade, which 
never could have been equalled as a silent film, for the very lines, 
spoken so well by all the players, especially Diana Wynyard, Clive 
Brook, and young Frank Lawton, were something that could never 
have been handled in mere text. I saw a photograph of the play 
Calvalcade, as it was done on the Drury Lane stage, and Winfield 
Sheehan's masterful production was greatly superior to the flesh-and- 
blood performance, in cast and scenic effects. Thinking up the sev- 
eral reels of titles that would have been needed to give anything like 
an adequate idea of the conversation between the characters may 
cause persons to reflect upon the fact that with even the skeletonized 
text of silent films the titles took about one-fifth the length of the 
picture. Then, too, one might miss a few words, through somebody's 



430 MORDAUNT HALL [j. s. M. p. E. 

passing his seat, and much of what went on for several scenes would 
be rather obscure. Nowadays even though one's view of the screen 
may be shut off temporarily, the voices are heard, and in any good 
film there is never the least doubt concerning what has happened. 

The dubbing in foreign countries of American productions appears 
to have been more satisfactory than was expected. Then there are 
also those superimposed titles, which, to my mind are even more 
satisfactory than giving a player a voice that does not belong to him. 
Anent dubbing, there have been films dealing partly with thugs and 
crooks who were supposed to be British. Occasionally a dialect was 
attempted, but more often than not it seemed as though all the gentry 
involved were old Oxonians. However, if the dubbing is done with 
greater care, it does help to give the producers additional financial 
returns. Few such offerings are shown in this country, as the dis- 
tributors of foreign productions here usually prefer the superimposed 
title to the dubbed voices. 

The most successful British films put on over here have been, The 
Drey fuss Case, The Private Life of Henry VIII, Catherine the Great, and 
Rome Express. The Henry VIII production captured popular fancy 
and was the inspiration of many a humorous picture in the periodi- 
cals. Although it was directed by the Hungarian, Alexander Korda, 
it was made in England and the title r61e was played by that brilliant 
British actor, Charles Laughton, who obviously could never have done 
so well by pantomime in a silent film. 

It must be agreed, however, that notwithstanding certain unin- 
spired uses of the screen's audibility, there have been many instances 
of really fine works. And it is to be hoped that the producers will 
do more to lead the public rather than give even as many instances 
of catering to the moronic fancy as they have in the past. A well- 
told story with imaginative ideas invariably appeals to all manner of 
men. 

DISCUSSION 

CHAIRMAN CRABTREE: Have you ever noticed that between the end of one 
reel and the beginning of another, a little black circle appears upon the screen? 
It often has a white circle around it. That is the signal for the projectionist to 
push the lever to change over to the next reel. Have you ever noticed those 
spots, and do you think the audience notices them? 

MR. HALL: No, I don't think I have, really. I have noticed where there have 
been cuts and so forth, but I believe that even those things pass unobserved by 
the spectator as he becomes really absorbed in the story. 



May, 1935] THE AUDIBLE PICTURE 431 

CHAIRMAN CRABTREE: You have viewed hundreds of pictures; have you felt 
any desire to see a larger picture than what you have seen? 

MR. HALL: I don't believe so; because when they tried it with the Grandeur 
screen, the voices were very poor when several characters were on the screen 
The dialog seemed to be coming from the corners of the screen and not from the 
actors. 

CHAIRMAN CRABTREE : We have seen recently some very excellent three-color 
pictures, La Cucaracha, for example. Color pictures up to within the past year 
have been produced largely by two-color processes, so that we could not reproduce 
yellow, purple, rose, and so on. The colors were not what we should call natural 
colors. But now with a three-color process available and perfected to quite a 
wonderful degree, what is your reaction, and what do you think the public's re- 
action is to these improved pictures? 

MR. HALL: I have seen only the one. I do think the color improves the Walt 
Disney subjects a great deal; now the Silly Symphonies are far superior to the 
Mickey Mouse features. But La Cucaracha seemed to me to be very dark and 
somber in spots. In most of the color pictures I have seen, there seems to be 
too much of a desire to show color and flash it before you. 

MR. ROGERS: Have you any criticism as to the way the light is applied in mak- 
ing motion pictures? My own criticism is that the details are too much enhanced. 
In some scenes even the darkest corners are shown with great detail. I should 
think that very often "atmosphere" could be created by lighting in a more natu- 
ral manner the Rembrandt effect, for example leaving out the unimportant 
details and letting the light also have a dramatic effect upon the picture. 

MR. HALL: Such extremely artistic pictures are all very well now and then, 
but whether they would have an appeal commercially, I do not know. 






A GLOSSARY OF COLOR PHOTOGRAPHY* 



The following glossary which has been prepared by the Color Committee contains 
some two hundred terms used in connection with color photography. Although, for 
convenience, all the listed terms are arranged alphabetically, it is really made up of 
two groups of material. One group consists of technical terms useful in the art. 
This group is composed for the most part of terms for which satisfactory definitions 
are not easily found in the text-books and dictionaries of the fundamental sciences, or 
in the general glossary of the Society.** It is not, therefore, intended as a complete 
technical vocabulary of the subject. The other group (marked by italicized type) 
contains proprietary names of special processes and equipment. The Committee 
admits to some uncertainty in connection with this portion. Some processes that 
are here described may be of no lasting importance; possibly other processes which 
have been omitted should have been included. In the case of some of the proprietary 
processes, complete, accurate information is not available; in the case of others the 
nature of the marketed product is changing from time to time. 

Absorption Band (of a color filter) A dark zone in a spectrum result- 
ing from the failure of a color-filter to transmit light of wave- 
lengths corresponding to the band. 

Acid Dyes Dyes in which the color resides in the negative ion 
(anion). Commonly, salts of colorless inorganic bases with 
colored or potentially colored organic acids. 

Additive Mixture See ADDITIVE SYNTHESIS. 

Additive Process A process for reproducing objects in natural 
colors by means of the principle of additive synthesis. Usually, 
black-and-white positives, printed from negatives taken 
through the primary color filters, are projected or viewed in 
register by means of light beams of the primary colors. 

Additive Synthesis The formation of a color by mixing light of 
two or more other colors. Any color may be formed by mixing 
light of three primary colors in the proper proportions. Some 
colors may be formed by mixing light of two other colors. 

Ag/acolor Process A 16-mm. adaptation of the lenticulated film 
principle. (1932) 

* Received December 19, 1934. 

** /. Soc. Mot. Pict. Eng., XVII (Nov., 1931), No. 5. p. 819. 
432 



GLOSSARY OF COLOR PHOTOGRAPHY 433 

Angstrom Unit A unit of length generally used for specifying the 
wavelength of radiation, especially light and radiant energy 
of wavelengths shorter than light. Numerically equal to 
0.0000001 mm. (10- 7 mm.). The unit more frequently used in 
colorimetrics is the millimicron. 

Aniline Dyes A term broadly applied to synthetic dyes derived 
from aniline or other coal-tar products. 

Artificial Daylight Light (visible radiation) having the same (or 
nearly the same) spectral composition as direct solar radiation 
plus skylight in practice produced by selectively absorbing 
some components of the light emitted by artificial sources. 

Autochrome Process A process for three-color additive photog- 
raphy, plates for which are made by Lumiere. The plates 
carry an irregular mosaic screen of red, green, and blue-violet 
starch grains with a panchromatic emulsion over-coating. 
(1907) 

Basic Dyes Dyes in which the color resides in the positive ion 
(cation). Commonly, salts of colored organic bases with 
colorless acids. 

Beam-Splitter An optical system so arranged as to reflect or 
transmit two or more portions of a light-beam along different 
optical axes. Such a device is frequently used in the produc- 
tion of color-separation negatives. 

Bichromated Gelatin Gelatin sensitized to light by the incorpora- 
tion of a soluble bichromate, usually ammonium or potassium 
bichromate. 

Bipack A unit consisting of two superposed films or plates sensi- 
tive to different portions of the spectrum, and intended to be 
exposed one through the other. 

Biprism A prism having a principal section which is a triangle 
with a very obtuse angle and two equal acute angles, sometimes 
used as a beam-splitter. 

Black Incapable of reflecting light. 

Black Body 1. A body which when heated radiates ideally ac- 
cording to fundamental physical laws (i. e., Wien radiation 
law) relating energy, frequency, and absolute temperature. 
The properties of incandescent tungsten or carbon approxi- 
mate those of a black body. 2. A body which absorbs all 
light incident upon it. 

Bleach v. t. To remove the color by chemical means; in photo- 



434 GLOSSARY OF COLOR PHOTOGRAPHY [j. s. M. P. E. 

graphy to remove, by chemical action (usually oxidation), 
the silver of an image. An image thus treated may be re- 
stored by suitable means generally leaving the gelatin film 
toned and/or tanned, n. A chemical reagent used for bleach- 
ing. 

Bleach-Out Process A process for making color-prints from a color- 
transparency, by use of a support coated with a mixture of 
dyes, each of which is capable of being decolorized by exposure 
to light in a different portion of the spectrum. 

Bleeding of Color The diffusing of dye away from a dye-image; 
most noticeable where dark areas adjoin light areas in a picture. 

Brewster Process A subtractive two-color process utilizing a 
double-coated negative film. A colored negative is printed 
on double-coated positive film, and the final silver images 
are bleached and dyed. (1914) 

Brightness The light (luminous power) per unit area, per unit 
solid angle, emitted from a surface in a given direction; the 
candle-power per unit area. 

Brilliance The characteristic of a color that expresses the inten- 
sity of the sensation. 

Busch Process An additive two-color process. The negative is 
produced by running 35-mm. film horizontally through the 
camera. Twin lenses form a pair of images upon a single frame 
area; image pairs are superposed when projected. (About 
1928) 

Carbon Printing A process for making prints in one or more colors 
by exposure of a bichromated and pigmented gelatin tissue to 
produce local insolubility of the gelatin, followed by the de- 
velopment of a relief image through the solvent action of 
warm water. 

Carbon Transfer Process A process in which a relief image, pro- 
duced in carbon printing, is transferred to another support 
from the one upon which it was developed. 

Chemical Toning The process of converting the silver of a photo- 
graphic image into a colored substance, or replacing it by a 
colored substance through the use of chemical reagents which 
are not dyes. 

Chromatic Aberration A defect of a lens resulting in a difference 
in focal length for light of different colors. 



May, 1935] GLOSSARY OF COLOR PHOTOGRAPHY 435 

Chromaticity The quality of colors that embraces hue and satura- 
tion but excludes brilliance. 

Chromoscope 1. A viewing-device for obtaining superposed 
images of color separation positives. 2. A type of colorime- 
ter using colors produced by the rotary dispersion of quartz 
as standards. 

Cine Color Process A sub tractive three-color process. Nega- 
tives are made with a beam-splitter camera using a single film 
and a bipack. Double-coated film is used for the red (dye 
tone) and blue (iron tone) images. The third (yellow) image 
is added to the film from a matrix by imbibition. 

Colloidal Dyes Dyes the particles of which are submicroscopic 
in size, but larger than molecules or ions. 

Color 1. The general name for all sensations (other than those 
related to spacial distribution) arising from the activity of 
the eye and its attached nervous system. Examples of color 
are the sensations red, yellow, blue, black, white, gray, etc. 
2. More loosely, as above but excluding the black, white, and 
gray sensations. 

Color Analyzer 1. A colorimeter. 2. An instrument used to 
determine the relative brightness of light of different wave- 
lengths reflected or transmitted by a substance or emitted by 
a source. 

Color Balance The adjustment of the intensities of printing or 
viewing colors (of a color picture) so as to reproduce properly 
the scale of grays. 

Color Blindness An ocular defect resulting in failure of the eye 
to distinguish between chromatic colors. In total color blind- 
ness all colors appear as grays; the more usual partial color 
blindness (dichromatism) is marked by inability to distinguish 
between certain pairs, as, for instance, red and green. 

Color Contrast See CONTRAST, COLOR. 

Color cr ajt Process A two-color sub tractive process of cinematog- 
raphy. The negative is made by a beam-splitter or by a bi- 
pack method; the positive is on double-coated film. Print 
images are dye toned with the aid of an iodide mordant. 
(About 1929) 

Color Developer A substance or mixture of substances capable 
of reducing silver halides with the simultaneous production 



436 GLOSSARY OF COLOR PHOTOGRAPHY [J. S. M. p. E. 

of an insoluble colored oxidation product in the regions of the 
silver deposit. 

Color Filter See FILTER. 

Colorimeter An instrument used for measuring color. The mono- 
chromatic colorimeter operates according to the principle that 
any color sensation may be matched by a pure spectral hue mixed 
with white, or by adding a pure spectral hue to the unknown 
color to produce white. The trichromatic colorimeter operates 
according to the principle that any color sensation may be 
matched by the addition in the proper proportion of three pri- 
mary colors, viz., red, green, and blue. 

Colorimetric Purity (of a color) The ratio of the luminosity of the 
dominant wavelength to the total luminosity of the color 
being measured. 

Color Index A publication of the Society of Dyers and Colorists 
(British), listing practically all dyestuffs in commercial use. 

Color Match The condition resulting when samples of light from 
two or more sources produce identical color sensations. 

Color Mixture Curves See COLOR SENSATION CURVES. 

Color Negative A negative record of the color values of the original 
object. 

Color Photography A process in which an attempt is made to re- 
produce objects in their natural colors by photographic means. 

Color Positive A positive photographic (print) record of color 
values. 

Color Saturation See SATURATION, COLOR. 

Color Screen 1. A color filter. 2. A surface bearing a mosaic, 
either regular or irregular, of minute, juxtaposed, trans- 
parent elements of the primary colors; used in a screen-plate 
or screen-film process of color photography. 

Color Sensation Curves (Excitation Curves) Curves based upon 
the response of the normal human eye, showing the relative 
excitations of the three elementary sensations, according to 
the Young-Helmholtz theory of color vision. 

Color Sensitivity, Photographic The sensitivity of a photographic 
material to light of various portions of the spectrum. 

Color Separation The obtaining of separate photographic records 
of the relative intensities of the primary colors in a subject 
in such a manner that a photograph in natural colors can be 
produced therefrom. 



May, 1935] GLOSSARY OF COL0R PHOTOGRAPHY 437 

Color Specification A description of a color made in such a way 
that the color sensation may be duplicated. This may be 
done either with the aid of a color analyzer or by the use of 
certain visual color matching devices, such as colorimeters or 
color comparators. 

Color Temperature (of a source) The temperature (expressed on 
the absolute scale) at which a black body radiator will visually 
match the color of the source. 

Color Transparency A color photograph upon a glass or film sup- 
port to be viewed or projected by transmitted light, as dis- 
tinguished from a color photograph on paper or other opaque 
white support to be viewed by reflected light. 

Color Tree A graphical method for specifying color sensation. 
Brilliance, hue, and saturation are presented in three dimen- 
sions. 

Color Triangle A graphical method of specifying hue and satura- 
tion. The three primary colors are represented at the apexes 
of a triangle and white at its center. 

Complementary After-image A sensation caused by ocular fatigue 
characterized by the persistence of an image of the color com- 
plementary to that of the original stimulus. 

Complementary Colors Two colors of light, which, when added 
together in proper proportions, produce the sensation of white 
or gray. Also, two colors of dye or pigment, which, when 
superposed in proper concentrations, produce a gray. 

Cones One of the two chief light-sensitive elements of the retina, 
frequently regarded as the seat of color vision. See RODS. 

Continuous Spectrum A spectrum, or section of a spectrum, in 
which radiations of all wavelengths are present; opposed to 
line spectra, or band spectra. 

Contrast, Color The ratio of the intensities of the sensations 
caused by two colors. Sometimes the logarithm of this ratio. 

Daylight Total radiation from the sky and sun. For standardiza- 
tion of spectral quality, measurements are made at noon. 
The quality of daylight matches approximately that of a 
black body at 6500 degrees Kelvin. 

Density The logarithm to the base 10 of opacity (for transparent 
materials). The logarithm of the reciprocal of the reflecting 
power (for reflecting materials) . 

Desensitization Treatment of a photographic material, as with 



438 GLOSSARY OF COLOR PHOTOGRAPHY [j. s. M. p. E. 

a solution of a suitable dye, to reduce its sensitivity to subse- 
quent light exposure without destroying the developability of 
a previous exposure. 

Developed Color Images Colored photographic images produced 
by direct development. 

Dichroic Pertaining to the property of certain crystals of showing 
different colors when viewed in different directions by trans- 
mitted light; or pertaining to the property of some solutions 
of varying color with layer thickness or concentration. 

Dichromatism See COLOR BLINDNESS. 

Differential Hardening (of gelatin) The production of an image 
in gelatin in a manner such that the hardness is proportional 
to the original silver density of the image; or, in other cases, 
to the amount of light which has fallen upon a specially 
treated gelatin coating. 

Dominant Wavelength In a system of monochromatic colorime- 
try the wavelength, the hue of which matches the hue of the 
color being measured. 

Double-Coated Film Film having a sensitive emulsion on both 
sides of the base, the emulsions or the base often being im- 
pregnated with a dye which prevents the penetration of ac- 
tinic light to the opposite emulsion when exposing either 
one of them. 

Double-Image Prism A prism so designed that with a lens it will 
form two images of an object; a beam-splitter. 

Dufaycolor Process A regular mosaic screen-plate process for three- 
color additive cinematography . (1931) 

Dufay Process A regular mosaic screen-plate process using four 
constituent colors. (1908) 

Dupack Process A process using a combination of a green-sensi- 
tive and panchromatic film sold by du Pont for making two- 
color motion picture negatives. The green-sensitive film bears a 
red filter layer upon its emulsion surface. The two films are 
run through the camera with their emulsion sides in contact. 
Exposure is made through the base of the green-sensitive film. 
(About 1931) 

Duplex Color-Plates Similar to the Paget screen-plate. The 
regular mosaic screen and the sensitive emulsion are on sepa- 
rate plates. (About 1927) 

Dye Density 1. The logarithm to the base 10 of the visual opacity 



May, 1935] GLOSSARY OF COLOR PHOTOGRAPHY 

of an area in a finished dye image. 2. The density of a single 
component of a two- or three-color print as measured by light 
of the complementary color. 

Dye Mordanting Broadly, the process of fixing a dye to a sub- 
stance for which it has no affinity by means of a second sub- 
stance which has an affinity both for the dye and for the first 
substance. More especially, in color photography, the treat- 
ment of a silver image so as to replace it in whole or in part 
with a substance having an affinity for dyes. 

Dye Toning The process of affixing a dye to a silver image or of 
replacing a silver image by a dye image through the action of 
mordants. 

Effective Wavelength See DOMINANT WAVELENGTH. 

Elementary Colors See PRIMARY COLORS. 

Elements (of a screen-plate or lenticular color-film) The indi- 
vidual filter particles of a color-screen, or the minute lenses of 
a lenticular film. 

Embossing v. t. The process of impressing minute lens elements 
upon a film base to produce a lenticular color-film, n. The 
lens elements collectively. 

Equality of Brightness (of colors) The state in which two colors 
have equal visual luminosity. 

Etch To dissolve portions of a surface not protected by a resist, 
as in making a halftone plate on copper or zinc ; also to remove 
differentially hardened gelatin from an image. 

Excitation Curves See COLOR SENSATION CURVES. 

Farbst off-Tab ellen (Lehmann and Schultz) A listing of commer- 
cial dyestuffs similar to that of the Color Index. 

Filter A light-transmitting material (or liquid solution in a cell) 
characterized by its selective absorption of light of certain 
wavelengths. A so-called "neutral gray" filter absorbs light 
of all wavelengths to which the eye is sensitive to approximately 
the same extent and so appears without hue. 

Filter Cut The wavelength or spectral region at which the absorp- 
tion of the filter varies rapidly with changing wavelength. 

Filter Factor (Filter Ratio) The ratio of the exposure required to 
produce a given photographic effect when a filter is used to 
that required without the filter. Many considerations, such as 
color-sensitivity of the emulsion, quality of radiation, and time of 
development influence the filter factor. 






440 GLOSSARY OF COLOR PHOTOGRAPHY [j. s. M. P. E. 

Filter Overlap The spectral region in which two or more filters 
transmit light effectively. 

Filter Ratio See FILTER FACTOR. 

Finlaychrome See Finlay Process. 

Finlay Process A regular mosaic screen-plate process of color 
photography utilizing either a screen separate from a panchro- 
matic plate (1929) or coated upon the same plate. The latter 
type is known under the trade-marked name, "Finlaychrome." 
(1931) 

Flicker Photometer An instrument in which two colors are pre- 
sented alternately to the eye. Above a certain minimum fre- 
quency, equally brilliant colors show no flicker such as is 
shown by colors of different brilliance. 

Fraunhofer Lines Definitely located absorbed lines in the solar 
spectrum. Certain of the lines are named by letter, their posi- 
tions being used as references of position in the spectrum. 

Fringe A defect in a color picture resulting from lack of registra- 
tion of the component images. A fringe may be caused by 
parallax, error in printing registration, or by movement in 
the object which has taken place between the exposure of 
color-separation negatives. 

Fundamental Colors See PRIMARY COLORS. 

Caspar Process A three-color subtractive motion picture process. 
Prints are made on film coated with three emulsion layers sensi- 
tized to three different spectral regions. In each emulsion is 
incorporated a dye which is destroyed in a bleach bath to a de- 
gree controlled by the silver image density. (1934) 

Gaumont Tri-Color Additive Process An additive method of three- 
color cinematography using a triple lens system both in the 
camera and in the projector. The frames are of standard 
(silent) width and three-fourths the standard height. (1912) 

Gelatin Filter A filter in which gelatin is used as the vehicle for 
the absorbing material. 

Gray Filter See FILTER. 

Gray Key Image x An image of neutral color occasionally printed 
in register with the images in tri-color inks or dyes. In the 
imbibition process, the gray key image is sometimes developed 
on the printing material by the ordinary photographic method. 

Half Silvered Mirror See MIRROR, SEMI-TRANSPARENT. 

HandscKiegel Process A process of applying color to local areas 



May, 1935] GLOSSARY OF COLOR PHOTOGRAPHY 441 

of black-and-white prints by imbibition, using one or more 
dyed matrices. 

Harriscolor Process A two-color subtractive process of cinematog- 
raphy. Prints from color-separation negatives are made on 
single-coated film printed first through the back, processed, 
and blue-toned with iron. The residual emulsion on the front 
is subsequently printed, processed, and red- toned. (1929) 

Herault Trichrome Process An additive three-color process for 
cinematography. The three-color print, consisting of suc- 
cessive red, green, and blue dye-tinted frames, is projected 24 
frames per second in a non-intermittent projector. (About 
1929) 

Heterochromatic Photometry The comparison of the intensity of 
light of different colors. 

Horst Process An additive three-color process in which the three 
images are exposed and later printed within one standard 
frame. (About 1929) 

Hue That attribute of certain colors in respect of which they differ 
characteristically from the gray of the same brilliance, and 
which permits them to be classed as reddish, yellowish, green- 
ish, or bluish. 

Hue Sensibility The sensibility of the eye to differences of hue. 

Hypersensitization The treatment of an unexposed photographic 
material by immersion in a solution, such as ammonia, to in- 
crease its sensitivity, principally to longer wavelengths. 

Imbibition A process for producing a dye-image by mechanical 
printing. A dyed relief or differentially tanned matrix of some 
substance such as gelatin is brought into intimate contact with 
a moist absorbing layer such as gelatin, the dye diffusing from 
the matrix to the absorbing layer. 

Imbibition Matrix A coating of gelatin or other colloid upon a 
support having an image capable of being dyed with water- 
soluble dye. See IMBIBITION. 

Interference Colors Colors resulting from the destruction of the 
light of certain wavelengths, and the augmentation of the 
light of others in a composite beam by interference. Colors 
of thin films and polarization colors of doubly refracting crys- 
tals in the polariscope are examples of interference colors. 

Isopaque Curve A line connecting a series of points of equal 
opacity. Such curves when applied to spectrograms may be 



442 GLOSSARY OF COLOR PHOTOGRAPHY [j. s. M. p. E. 

used to demonstrate the color-sensitivity of photographic 
materials. 

Joly Color Screen A regular mosaic screen-plate consisting of 
ruled lines. (1894-5) 

Keller *D or iari'Berthon Process See Keller-Dorian Process. 

Keller-Dorian Process -A three-color additive motion picture 
process. A banded tricolor filter is associated with the 
camera lens. The film support which faces the lens is embossed 
with small lens elements. Each lenticular element images the 
filter bands upon the emulsion surface. A filter of similar 
form is associated with the projection lens. (Pat. 1908-9; 
introduced 1925) 

Kinemacolor Process A two-color additive process involving the 
use of a rotary shutter of color-filters before the lenses of both 
camera and projector. (1906) 

KodacKrome Process A two-color subtractive process for still 
photography and 35-mm. motion pictures, devised by the 
Eastman Kodak Company. Prints are made upon double- 
coated film ; the positive is bleached with a tanning bleach and 
dyed with dyes which penetrate soft gelatin preferentially. 
(1915) 

Kodacolor Process A 16-mm. adaptation of the Keller-Dorian 
process. (1928) 

Kromogram Three transparent stereoscopic pairs of images which 
appear as a single color picture when viewed in a special view- 
ing device called the "Kromskop." (1894) 

Kromskop A special form of the chromoscope invented by F. E. 
Ives, utilizing two semi-transparent mirrors and suitable color- 
filters for exposing three images with one lens. Positives 
(Kromograms) printed from the images are viewed with a 
similar device. (1894) 

Lake A pigment formed by the combination of an organic dye 
with a metallic compound or another dye with which it forms 
an insoluble precipitate. 

Lenticulation Minute optical elements having the form of cylindri- 
cal or spherical lenses embossed into the support side of photo- 
graphic film. They serve in the process of analysis and syn- 
thesis of images in an additive color process. See Keller- 
Dorian Process. 

Leuco-Base A white or slightly colored substance which, upon 



May, 1935] GLOSSARY OF COLOR PHOTOGRAPHY 443 

oxidation, sometimes accompanied by reaction with an acid 
or base, yields a more highly colored dye. 

Light Radiant energy evaluated according to its capacity to pro- 
duce visual sensation. 

Light Restraining Dye A dye used for impregnating a light-sensi- 
tive emulsion to prevent the deep penetration of light during 
exposure. 

Light-Splitter See BEAM-SPLITTER. 

Lignose Process An irregular mosaic three-color process applied 
to roll film and film pack. (1927) 

Line-Screen Process A color-screen process in which the screen 
is formed by a regular pattern of ruled lines. 

Lippmann Process A process of direct color photography based 
upon the interference of light. An exceedingly fine-grained 
panchromatic emulsion is exposed in intimate contact with a 
metallic (mercury) mirror. A standing-wave pattern is pro- 
duced throughout the depth of the emulsion layer, the silver 
being reduced in the anti-nodal planes, thus forming a sys- 
tem of reflecting laminae. The plates are viewed by reflected 
light. (1891) 

Magnachrome Process A two-color additive process of color cine- 
matography. Half the normal picture height is used for each 
of the pairs of pictures. 

Magnacolor Process A two-color sub tractive process for cine- 
matography. Bipack negative and double-coated positive 
films are used. (1930) 

Maxwell Experiment The first demonstration of the principle of 
additive synthesis with color-separation negatives. Clerk 
Maxwell and Thomas Sutton in 1861 produced a set of four 
plates and projected them in register before an audience. 

Maxwell Primaries The colors red, green, and blue-violet used by 
Maxwell to demonstrate the application of the Young-Helm- 
holtz theory to color photography. 

Micron A unit of length equal to 0.001 mm. (10~ 3 mm.). Used 
frequently to designate the wavelength of radiant energy in 
the infra-red region. 

Millimicron A unit of length equal to 0.000001 mm. (10~ 6 mm.). 
This is the unit usually used in colorimetrics in expressing the 
wavelength of radiant energy. 

Minus Color The color which is complementary to the color that 



444 GLOSSARY OF COLOR PHOTOGRAPHY [J. s. M. P. E, 

is named; for example, minus red is a color complementary 
to red. 

Mirror, Semi-Transparent A mirror uniformly coated with reflect- 
ing material in such a manner that part of the light incident 
upon it is reflected, the other part passing through the surface. 
A type of beam-splitter. 

Moire* Effect A "watered" pattern produced when two or more 
screens bearing a system of fine regular lines or similar pat- 
tern are superposed nearly but not exactly in register. 

Monochrome A picture executed in a single color. 

Mordanting See DYE MORDANTING. 

Morgana Process A two-color additive process of color cinematog- 
raphy (for 16-mm. reversal pictures). In the projector, 
the film is moved two frames forward, one backward, and so 
on. Effective camera and projection speed is 24 frames per 
second, although the special projector movement produces 72 
alternations per second. (1932) 

Mosaic Screen Plate A color screen plate. 

Motion Fringe A fringe of color occurring at the edge of images 
when the color-separation negatives are taken at different in- 
stants. 

Multicolor Process A two-color subtractive 35-mm. cinematogra- 
phic process. The negative is made with a bipack. The col- 
ored print is made on double-coated film. (1929) 

Neutral Color Gray; achromatic; possessing no hue. 

Neutral Filter See FILTER. 

Neutral Wedge A wedge composed of a neutral (gray) absorbent 
material. 

Orthochromatic 1. Characterizing the equivalence between the 
photographic effect of various colors upon a photographic ma- 
terial and the physiological effect upon the eye. 2. By usage, 
characterizing a photographic material sensitive to all colors 
except red. 

Paget Color Screen Plate A regular mosaic color screen plate 
(1912), available commercially since 1929 as the Finlay plate. 

Panchromatic Characterizing a photographic material sensitive 
to all colors of the visible spectrum. 

Parallax The apparent displacement of an object as seen from two 
different points. 

Pathechrome Process A cinematographic process in which color is 






May, 1935] GLOSSARY OF COLOR PHOTOGRAPHY 445 

applied to a black-and-white print through a celluloid film 
stencil. (1928) 

Photochromoscope See CHROMOSCOPE, Kromskop. 

Photocolor Process A two-color subtractive process using a twin 
lens camera and dye-toned prints on double-coated film. 
(About 1930) 

Pigment An insoluble colored material in finely divided form. 

Pilney Process A two-color subtractive cinematographic process. 
(1930) 

Pinachrome Process A printing process based upon the use of leuco- 
bases which oxidize upon exposure to light, yielding color images 
which are assembled by superposition. 

Pinatype Process A subtractive three-color process for still pic- 
tures based upon the differential staining action of certain dyes 
for hard and soft gelatin. (1906) 

Primary Colors Three colors, which, when mixed in the proper 
proportions, can be used to produce all other colors. The 
three colors most commonly used are red, green, and blue- 
violet. 

Prismatic Spectrum A spectrum formed by a prism. 

Purkinje Effect A shifting of the visibility curve to shorter wave- 
lengths at decreased intensities, e. g., as intensity is decreased 
object colors may change from reddish to bluish. 

Quality (color) CHROMATICITY. 

Quality of Radiation An expression which refers to the spectral 
composition of the radiation. In both photography and in 
the viewing of colored pictures the quality of the radiation 
used is important. 

Ratio Diaphragm Cap A mask placed over a banded tricolor filter 
shaped to permit a predetermined ratio of the different colors 
of light to pass through a filter for lenticulated film color photog- 
raphy. 

Ratiometer Any device used to test the actinic equality of dif- 
ferently colored lights transmitted to the photographic ma- 
terial in making color separation negatives. 

Raycol Process A two-color additive process of cinematography. 
The image pairs are exposed ( x /4 standard size) on each frame 
and disposed in diagonal corners of the frame. The image pairs 
from contact positives are superposed by a suitable optical 
system. (1930) 



446 GLOSSARY OF COLOR PHOTOGRAPHY [j. s. M. p. E. 

Register v. t. To cause to correspond exactly; to adjust two or 
more images to correspond with each other. Such correspon- 
dence may be required either in printing or in projection, n. 
A state of correspondence. 

Relief Process Any color process in which relief images are pro- 
duced, for the purpose of matrix printing. 

Resist A coating used to protect certain portions of a surface 
upon which an image or design is to be produced by etching, 
dyeing, or other chemical or physical treatment. 

Rods One of the two chief light-sensitive elements of the retina. 
See CONES. 

Saturation, Color The degree of freedom of a color from admixture 
with white. 

Saturation Sensibility The sensibility of the eye to saturation. 

Screens, Color See COLOR SCREEN. 

Screen-Plate Process See COLOR SCREEN. 

Sennettcolor Process A sub tractive cinematographic process using 
a bipack negative and a double-coated film for the print. 
(1930) 

Sensation Curves See COLOR SENSATION CURVES. 

Sensitizers Materials, usually dyes, used to increase the sensi- 
tivity of photographic emulsions to light of various wave- 
lengths. 

Separation, Color See COLOR SEPARATION. 

Sirius Process A two-color subtractive cinematographic process 
in which alternate frames of the negative are exposed with the 
aid of a beam-splitter, and the positive print is made upon 
double-coated film. (1929) 

Soft Gelatin Process A process in which there is preferential 
dyeing of soft gelatin portions of the image. See Pinatype. 

Spectral Composition (of radiation) The specification of the rela- 
tive energy at different wavelengths of radiation emitted by a 
source, or reflected or transmitted by a material ; usually shown 
graphically as a spectral distribution curve. 

Spectral Distribution Curve See SPECTRAL COMPOSITION. 

Spectral Energy Curve See SPECTRAL DISTRIBUTION CURVE. 

Spectral Sensitivity The sensitivity of a light-sensitive material 
(or instrument, such as a photoelectric cell) to radiation of 
various wavelengths. 

Spectral Transmission (of a filter) The extent to which a filter will 



May, 1935] GLOSSARY OF COLOR PHOTOGRAPHY 447 

transmit radiation of different wavelengths. Shown graphi- 
cally as transmission, opacity, or density plotted against wave- 
length. 

Spectrogram A photograph of a spectrum. See WEDGE SPECTRO- 
GRAM. 

Spectrophotometer A spectroscope with a photometric attachment 
used to determine the relative intensity of two spectra or 
spectral regions. 

Spectroscope An instrument for forming a spectrum and measur- 
ing wavelengths in various regions throughout the spectrum. 

Spectrum An image of a source formed by light or other radiant 
energy through the medium of an optical device which re- 
fracts or diffracts the radiation of different wavelengths to dif- 
ferent degrees. Throughout the visible spectrum of a con- 
tinuous light source the spectrum appears as a number of 
juxtaposed areas of color varying from red to violet; e. g., 
the rainbow. 

Spicer'Dufay Process See Dufaycolor Process. 

Splendicolor Process A three-color sub tractive process in which 
the three-color separation records are printed as follows: 
blue record upon one side by iron toning, and the yellow and 
red as successive color layers upon the opposite side by dyed 
bichromate methods. (1928) 

Subtractive Primaries The three printing colors used in a three- 
color subtractive process; usually named magenta (minus 
green), blue-green (minus red), and yellow (minus blue). 

Subtractive Process A process of reproducing objects in natural 
colors using a restricted number of primary component colors 
in which the composite image is produced by passing a single 
beam of white light successively through two or more layers 
of colored images, each of which absorbs one region of the spec- 
trum which is passed by the other layers. 

Tanning Developers Solutions which cause hardening, or which 
render insoluble, the gelatin of an emulsion in proportion to 
the amount of latent image converted into silver. 

Technicolor Process A trade-name applied to various types of 
subtractive cinematographic color processes. (About 1915) 
At one time marketed as a two-color relief process; more re- 
cently as a three-color imbibition process. 



448 GLOSSARY OF COLOR PHOTOGRAPHY [j. S. M. p. E. 

Three-Color Process Any process, either additive or subtract! ve, 

for producing photographs using three primary colors. 
Tinctorial Power (of a dye) The reciprocal of the concentration 

necessary to produce a given amount of absorption in a layer 

of given thickness. 
Tintometer An instrument for estimating or specifying color by 

comparison with a graded series of standard colors. 
Toning See CHEMICAL TONING, DYE TONING. 
Transfer Process A process in which an image, usually dyed or 

pigmented, is transferred from one surface to another. 
Trichromatic Process See THREE-COLOR PROCESS. 
Tricolor Filter 1. A composite filter containing areas of three 

primary colors; 2. A single filter of one of three primaries. 
Tripack Process A process of exposing three films (or plates) 

simultaneously, in which the films are arranged as a pack so 

that the outer films (or interposed filters) transmit certain 

portions of the light to expose the following layers, (cf. Bi- 

PACK) . 
Two-Color Process Any process, either additive or subtractive, 

for producing photographs using two colors. 
Utocolor Process A three-color subtractive transfer process using 

the bleach-out method for making a color print by printing 

from a color transparency. It depends upon the bleaching 

property of certain wavelengths for certain dyes. (1895) 
Visibility (of radiation) The ratio of the luminous flux (lumens) 

to the corresponding energy flux (watts). 
Visibility Curve A graphical representation of the relation between 

visibility (expressed relatively) and wavelength. This curve 

has a maximum in the green at 555 m/z. 
Vitacolor Process An additive two-color cinematographic process 

similar to Kinemacolor. (1930) 
Warner*Powrie Process A three-color regular line-screen process. 

(1905) 
Wedge An optical device composed of absorbing material in 

which the transmission varies progressively from point to 

point. vSuch a device may cause a variation in either hue or 

intensity, or both. 
Wedge Filter See WEDGE. 
Wedge Spectrogram A spectrogram produced by photographing 

a spectrum through a neutral wedge (sometimes an optical 



May, 1935] GLOSSARY OF COLOR PHOTOGRAPHY 449 

wedge), placed usually over the slit of the spectrograph. Such 
a spectrogram shows graphically the effective photographic 
sensitivity vs. wavelength for the photographic material and 
light source used. 

White Light Radiant energy which has a wavelength-intensity 
distribution such that it evokes a neutral gray (hueless) sensa- 
tion in the average normal eye. 

White Object An object of a color such that it reflects all wave- 
lengths of the visible spectrum equally; an object which if 
illuminated by white light will appear without hue to the 
average normal eye. 

W ratten Filters A widely used commercial brand of color-filters. 

Young-Helmholtz Theory (of color vision) An explanation of phe- 
nomena of color vision assuming three separate elements in 
the normal eye, each stimulated by a different section of the 
visible spectrum. 

Zoechrome Process A three-color subtractive process of color 
cinematography with a black-and-white key. In the camera 
every alternate frame is normally exposed ; on each remaining 
frame, three images are exposed through primary filters. The 
standard size image is printed first, and each of the color- 
images in succession is enlarged and superposed upon the first. 
Between successive printings, the film is varnished and re- 
coated with emulsion. Each image layer is dye-toned before 
the next layer is added. (1929) 

Zoetrope Process Probably the first color photography process 
using the rapid substitution of primary images before the eyes. 
(1869) 



COLOR COMMITTEE 
C. TUTTLE, Chairman 

P. D. BREWSTER L. A. JONES 

R. M. EVANS J. F. KIENNINGER 



NEW MOTION PICTURE APPARATUS* 



A MOTOR-GENERATOR FOR THE NON-ROTATING 
HIGH-INTENSITY ARC** 



The following descriptions of equipment were included among the presentations 
in the Apparatus Symposium at the New York Convention of the Society, Oct. 29- 
Nov. 1, 1934. 

The recent introduction of the d-c. high-intensity "Suprex" carbons has neces- 
sitated a new design of projection lamp, and has likewise presented an opportunity 
for developing a motor-generator for converting the alternating current of com- 
mercial power mains into direct current of the proper voltage and regulation for 
most efficiently utilizing the advantages of the new lamps and carbons. 

Until eight or ten years ago the motor-generator that was most widely used was 
the two-unit series type. Despite its shortcomings, it possessed the advantage 
that no ballast resistance was required. However, due to the increasing demand 
for more intense screen illumination, the reflector type of lamp and improved 
models of the original high-intensity lamps came to be quite generally adopted. 
At the same time, the new arcs demanded a more stable source of current than 
the series machine, and the two-unit multiple motor-generator became the ac- 
cepted standard. 

The two-unit multiple machine possesses the requisite flexibility and stability 
for operating lamps having voltages of 55 or more at the arc. But the advan- 
tages of flexibility and stability were achieved rather expensively, for the ballast 
resistance required to achieve them accounted for a loss of as much as 30 or 40 
per cent of the purchased power. In other words, somewhat less than half the 
purchased power is used, not for furnishing illumination for projection, but in 
uselessly and wastefully heating the ballast resistors. 

The three-unit "Stabilarc-Unitwin" motor-generator was designed to combine 
most of the advantages of the series and multiple types with as few as possible of 
their disadvantages. As shown in Fig. 1, the machine consists of two d-c. genera- 
tors, one for each arc, on either side of and direct-connected to an a-c. motor. 
Terminal leads for each of the three units are brought out separately into their re- 
spective conduit boxes, and all openings are guarded with removable perforated 
or louvre covers. The shaft revolves in four ball-bearings and ball-bearing flexi- 
ble couplings connect the a-c. motor with the d-c. generator unit on either side. 

The d-c. control panel is shown in Fig. 2. It includes two small field rheostats, 
one for each generator unit, a voltmeter, and a double-throw toggle switch for 
connecting the voltmeter in circuit with either generator unit. Fig. 3 shows the 



* Presented at the Fall, 1934, Meeting at New York, N. Y. 
** Automatic Devices Co., Allentown, Pa. 



4,50 



NEW MOTION PICTURE APPARATUS 



451 



comparative simplicity of the connections between the motor-generator, panel, 
and arcs. Each generator feeds directly into its own arc. Only the field and 
voltage circuits are connected to the control panel, so that the connecting wires 
may be of the minimum size allowable under local insurance regulations, thus 
effecting a substantial saving in wiring costs. 

Operating without load, the voltage of each generator is about 55 to 60, depend- 
ing upon the setting of the field rheostat. When the carbons are brought to- 




FIG. 1. Stabilarc-Unitwin motor-generator for d-c. Suprex carbon arcs. 

gether, so as to strike the arc, the voltage of the corresponding generator unit im- 
mediately and automatically drops to approximately 15 volts. The carbons are 
held together for a moment or two and then slowly separated, without the ex- 
plosive and sputtering action that is unavoidable when striking an arc fed from a 
multiple generator, due to the heavy short-circuit current that occurs before the 
arc is formed. This explosive action tends 
to destroy the crater and blow out the core 
of the carbon, apparently having the effect of 
hurling particles of incandescent carbon and 
copper against the mirror, and is responsible 
for excessive reflector repairs and replace- 
ments. By means of this "controlled strike," 
pitting of the lamp mirror is greatly reduced 
if not entirely eliminated, thus materially 
extending the useful life of the reflector. 

Referring again to Fig. 1, it will be noted 
that the three-unit combination is long and 
narrow, the proportions being similar to 
those of 3600 rpm. high-speed machines. 
However, as the synchronous speed of the 
Unitwin is only 1800 rpm., the peripheral speed is reduced accordingly, thereby 
very materially decreasing brush, wind, and other noises of rotation. Extremely 
quiet operation is the result, a very desirable feature when the motor-generator 
must be placed in or near the projection room. 

In the multiple type of motor-generator, the single-generator unit is depended 
upon to supply one or more arcs in operation while another arc is being started. 
During the change-over period, the large short-circuit current caused by striking 




FIG. 2. The d-c. control panel. 



452 



NEW MOTION PICTURE APPARATUS [J. S. M. P. E. 



the additional arc, added to the load already carried by the generator, usually 
amounts to two to three times the normal rating of the machine, causing a con- 
sequent strain upon the commutator, brushes, and other load-carrying elements. 
In contrast, a Unitwin generator unit never supplies power to more than one arc, 
and is therefore never overloaded. During change-over, or at any other time 
when the entire machine supplies two arcs, the burden of the double load is placed 
where it properly belongs on the rugged squirrel-cage motor without any cur- 
rent-carrying moving parts. 

For the same reason, since each generator unit supplies only one arc, striking 



Ter/n/na/ Board 
fns/de ftone/Sox. 



Suprex Carbon L amps. 




FIG. 3. 



Wiring connections between motor-generator, 
panel, and arcs. 



the second arc has no effect upon the first arc, and the illumination of the screen 
is uniform throughout the entire operation. 

The characteristics of the Unitwin are admirably suited to the requirements of 
the Suprex arc. Whether the same three-unit combination can be economically 
adapted to other types of arcs is a question that can not yet be definitely answered. 



THE DAVIDGE DEVELOPING APPARATUS* 

The effects of agitation of the developer in relation to the film during the opera- 
tion of developing motion picture film are fairly well known. The chief effects 
are (1) an increase in the rate of development, and (2) a partial offsetting of the 
effects of the reaction products of development as typified by the "Mackie Line." 

With any particular developing equipment, therefore, the developer should be 
sufficiently agitated, and the agitation should be non-directional so that the cur- 
rent impulses strike the emulsion surface at constantly changing angles, supplying 



* Roy Davidge Co., Hollywood, Calif. 



May, 1935] 



NEW MOTION PICTURE APPARATUS 



453 



new developer and displacing the by-products of development in such a manner 
as to avoid distortion in density where a heavily exposed area lies adjacent to one 
of less exposure. 

The Davidge developing apparatus (Figs. 4 and 5) consists of a wheel or rotor 
which acts as a carrier for the film and the spacing apron, the latter comprising a 
molded celluloid strip having buttons placed alternately at either side which 
contact the two surfaces of the interwound film at the perforation area. The 
buttons or protuberances are so elongated as to be incapable of entering the per- 




FIG. 4. Davidge developing apparatus: 1200-ft. unit. 



f oration, and also to act as baffles to disperse the current of developer from the 
space on the opposite side. 

The exposed film with the separator is wound upon the rotor in lengths up to 
1200 feet, and clipped at the outside. A spacing of approximately 3 /ie inch is at- 
tained between each successive turn on the rotor, and the combined film and 
separator are in no way attached to the rotor, being free to move and change rela- 
tionship to the impulse fins at every revolution. 

The rotor is made of non-corrosive tubular metal, and has a small fin which ex- 
tends outward from each spoke, 22 on each side for the 1000-ft. rotor. When 
wound, the unit is placed into the developing solution and revolved at approxi- 



454 NEW MOTION PICTURE APPARATUS [J. S. M. p. E. 

mately 20 revolutions a minute. The developing tank also has a baffling device 
which creates a side-to-side motion of the solution through the wheel. This 
motion, in turn, is deflected at constantly changing angles by the fins on the rotor. 
The rotation is mechanically reversed at intervals of one minute during de- 
velopment. 

Films developed by the system show readings that are identical on butted sensi- 
tometric strips placed throughout the wind. 



FIG. 5. Davidge 16-mm. developing unit. 




FIG. 6. Moviola film viewing machine, model C. 



May, 1935] NEW MOTION PICTURE APPARATUS 

MOVIOLA FILM VIEWING MACHINES' 



455 



The Moviola film viewing machine, model C, Fig. 6, is now, like model D, regu- 
larly equipped with a hand rheostat and hand switch for controlling the motor, 
in addition to the foot controller always furnished with this machine. The hand 
rheostat and foot controller are in series, and the hand switch short-circuits the 
foot controller. Therefore, when the hand switch is closed, the hand rheostat 
controls the speed of the motor; and when the hand switch is open, the hand rheo- 
stat controls the maximum speed of the motor attainable by the foot controller. 

Fig. 7 shows a Moviola film viewing and sound reproducing machine, model 
UDC, for use with composite film only, and specially equipped with a footage and 
frame counter. This machine is particularly suitable for checking prints in the 
exchanges, as each frame on a reel of film can be easily identified. 




FIG. 7. Film viewing and sound re- 
producing machine, model UDC. 



Moviola Company, Hollywood, Calif. 



456 



NEW MOTION PICTURE APPARATUS 



Fig. 8 shows the front view of a Moviola film viewing and sound reproducing 
machine, model UDSL. This machine is made to be used with a standard 35-mm. 
picture film and two separate sound films, one for standard 35-mm. film or 
ITVs-rnm. "split" sound film, and the other for 16-mm. sound film. It includes a 




FIG. 8. Film viewing and sound reproducing 
machine, model UDSL; for standard 35-mm. picture 
film with separate sound-track (full width or split) and 
an additional sound-head for 16-mm. sound-film. 

standard Moviola film viewing machine, model D, which contains a reversible, 
variable-speed motor, a standard Moviola sound-head, model SD, belt-driven by a 
reversible l / t -hp. constant-speed induction motor, and a Moviola sound-head, 
model SL, for 10-mm. film. Foot control is provided for each of the two motors, 
and a three-wire attachment cord (including a ground connection) is furnished. 



OFFICERS OF THE SOCIETY 

1935 






L. A. JONES 
Engineering Vice-President 



E. HUSE 
Executive Vice-P resident 



J. I. CRABTREE 
Editorial Vice-P resident 




H. G. TASKER 
President 






O. M. GLUNT 
Financial Vice-P resident 



A. N. GOLDSMITH 
Past-President 



W. C. KUNZMANN 
Convention Vice-President 
457 



458 



OFFICERS OF THE SOCIETY 




J. H. KURLANDER 

Secretary 



A. S. DICKINSON 
Governor 



T. E. SHEA 
Treasurer 




H. GRIFFIN 
Governor 




M. C. BATSEL 
Governor 




W. B. RAYTON 
Governor 




S. K. WOLF 
Governor 




L. W. DAVEB 

Chairman, Atlantic 

Coast Section 




G. F. RACKETT 

Chairman, Pacific 
Coast Section 



COMMITTEES OF THE SOCIETY OF MOTION PICTURE 
ENGINEERS, 1935 



(Corrected to April 20, 1935; additional appointments may be made at any 
time during the year as necessity or expediency demands.) 

COLLEGE COURSE IN MOTION PICTURE ENGINEERING 



H. T. COWLING 
A. C. HARDY 



W. H. CARSON 
O. O. CECCARINI 



D. P. BEAN, Chairman 
L. A. JONES 

COLOR 

J. A. BALL, Chairman 
C. H. DUNNING 
R. M. EVANS 

A. M. GUNDELFINGER 



G. E. MATTHEWS 
T. E. SHEA 



H. W. MOYSE 
A. WARMISHAM 



H. BUSCH 

A. S. DICKINSON 

G. C. EDWARDS 



EXCHANGE PRACTICE 

T. FAULKNER, Chairman 
A. Hi ATT 
J. S. McL/EOD 

HISTORICAL AND MUSEUM 



N. F. OAKLEY 
H. RUBIN 
J. H. SPRAY 



W. E. THEISEN, Chairman 
G. A. CHAMBERS J. A. DUBRAY 

W. CLARK G. E. MATTHEWS 

T. RAMSAYE 



JOURNAL AWARD 

A. C. HARDY, Chairman 



J. A. BALL 
O. M. GLUNT 



L. A. JONES 

E. A. WILLIFORD 



LABORATORY PRACTICE 

D. E. HYNDMAN, Chairman 
H. W. MOYSE, Vice-Chairman 
J. CRABTREE T. M. INGMAN 

R. M. EVANS M. S. LESHING 

E. HUSE C. L. LOOTENS 

R. F. MITCHELL 



J. M. NICKOLAUS 
W. A. SCHMIDT 
J. H. SPRAY 



459 



460 



COMMITTEES OF THE SOCIETY 



[J. S. M. P. E. 



Atlanta 

C. D. PORTER 



MEMBERSHIP ADMISSIONS 
M. W. PALMER, Chairman 
L. W. DAVEE P. A. McGuiRE 

P. J. LARSEN H. RUBIN 

J. L. SPENCE 

MEMBERSHIP AND SUBSCRIPTION 
E. R. GEIB, Chairman 

Detroit 

H. S. MORTON 

M. Ruben 
Boston 

T. C. BARROWS Hollywood 

J. S. CIFRE J. O. AALBERG 

J. R. CAMERON L. E. CLARK 

G. H. GIBSON 

C. W. HANDLEY 

E. HUSE 

F. E. JAMES 

R. G. LlNDERMAN 



Camden & Philadelphia 
H. BLUMBERG 
J. FRANK, JR. 



New York 

G. C. EDWARDS 

J. J. FINN 

G. P. FOUTE 

H. GRIFFIN 

W. W. HENNESSEY 

R. C. HOLSLAG 

M. D. O'BRIEN 

F. H. RICHARDSON 
H. B. SANTEE 
T. E. SHEA 
J. L. SPENCE 



Chicago 


G. A. MITCHELL 


J. H. SPRAY 


B. W. DEPUE 


P. MOLE 




J. H. GOLDBERG 


K. F. MORGAN 


Rochester 


S. A. LUKES 


G. F. RACKETT 


E. K. CARVER 


R. F. MITCHELL 








Minneapolis 


Travelling 




C. L. GREENE 


E. AUGER 


Cleveland 




K. BRENKERT 


R. E. FARNHAM 


Washington 


W. C. KUNZMANN 


J. T. FLANNAGAN 


N. GLASSER 


D. McRAE 


V. A. WELMAN 


F. J. STORTY 


O. F. NEU 






H. H. STRONG 




England 




A ustralia 


W. F. GARLING 


India 


H. C. PARISH 


D. McMASTER 


H. S. MEHTA 




S. S. A. WATKINS 


L. L. MISTRY 






M. B. PATEL 


A u stria 


France 




P. R. VON SCHROTT 


L. J. DlDIEE 


Japan 




L. G. EGROT 


T. NAGASE 




F. H. HOTCHKISS 


Y. OSAWA 


Canada 


J. MARETTE 




F. C. BADGLEY 






C. A. DENTELBECK 


Germany 


New Zealand 


G. E. PATTON 


W. F. BIELICKE 


C. BANKS 




K. NORDEN 




China 




Russia 


W. D. COOLEY 


Hawaii 


A. F' CHORINE 


R. E. O'BOLGER 


L. LACHAPELLE 


G. E. JACHONTOW 



MAY, 1935] 



COMMITTEES OF THE SOCIETY 



461 



V. C. ARNSPIGER 

D. P. BEAN 

E. W. BEGGS 
W. B. COOK 



G. A. CHAMBERS 
C. DREHER 
J. A. DUBRAY 



NON-THEATRICAL EQUIPMENT 
R. F. MITCHELL, Chairman 
H. A. DEVRY 
R. E. FARNHAM 
E. C. FRITTS 
H. GRIFFIN 

PAPERS 

J. O. BAKER, Chairman 
R. E. FARNHAM 
H. F. JERMAIN 
G. E. MATTHEWS 

PROGRESS 



R. HOLSLAG 

E. Ross 
A. SHAPIRO 
A. F. VICTOR 



W. A. MUELLER 

D. RIDGWAY 

E. O. SCRIVEN 



J. G. FRAYNE, Chairman 
M. ABRIBAT R. E. FARNHAM 

L. N. BUSCH E. R. GEIB 

A. A. COOK G. E. MATTHEWS 

R. M. CORBIN H. MEYER 

J. A. DUBRAY V. E. MILLER 

PROGRESS MEDAL AWARD 
A. N. GOLDSMITH, Chairman 
M. C. BATSEL C. DREHER 

J. CRABTREE W. B. RAYTON 

PROJECTION PRACTICE 
J. O. BAKER, Chairman 
J. J. FINN 
E. R. GEIB 
H. GRIFFIN 
C. W. HANDLEY 
J. J. HOPKINS 



R. F. MITCHELL 
G. F. RACKETT 

P. R. VON SCHROTT 

S. S. A. WATKINS 
I. D. WRATTEN 



J. O. AALBERG 
T . C. BARROWS 

F. C. CAHILL, JR. 
J. R. CAMERON 

G. C. EDWARDS 
J. K. ELDERKIN 



C. F. HORSTMAN 

P. A. McGuiRE 

R. MlEHLING 
M. D. O'BRIEN 

H. RUBIN 
V. A. WELMAN 
PROJECTION SCREEN BRIGHTNESS 

C. TUTTLE, Chairman 
G. A. CHAMBERS, Vice-Chairman 

J. O. AALBERG F. L. EICH J. H. SPRAY 

B. BERG C. W. HANDLEY A. T. WILLIAMS 

A. A. COOK W. F. LITTLE S. K. WOLF 

G. F. RACKETT 

ex offrcio 

J. O. BAKER, Chairman, Projection Practice Committee 
D. E. HYNDMAN, Chairman, Laboratory Practice Committee 



PUBLICITY 
W. WHITMORE, Chairman 



J. R. CAMERON 
J. J. FINN 



G. E. MATTHEWS 
P. A. McGuiRE 






F H. RICHARDSON 



462 



COMMITTEES OF THE SOCIETY 



M. C. BATSEL 
W. H. CARSON 
A. C. DOWNES 
P. H. EVANS 
R. E. FARNHAM 
C. L. FARRAND 
H. GRIFFIN 



M. C. BATSEL 
R. M. EVANS 
E. H. HANSEN 



W. C. KUNZMANN 

J. H. KURLANDER 



STANDARDS 

E. K. CARVER, Chairman 
J. A. DUBRAY, Vice-Chairman 
R. C. HUBBARD 
E. HUSE 
C. L. LOOTENS 
K. F. MORGAN 
N. F. OAKLEY 
G. F. RACKETT 
W. B. RAYTON 

SOUND 

P. H. EVANS, Chairman 
K. F. MORGAN 
W. A. MUELLER 
O. SANDVIK 

STUDIO LIGHTING 

R. E. FARNHAM, Chairman 
G. F. RACKETT, Vice-Chairman 
V. E. MILLER 
M. W. PALMER 
E. C. RICHARDSON 

ATLANTIC COAST SECTION 
L. W. DAVEE, Chairman 



C. N. REIFSTECK 
H. RUBIN 
O. SANDVIK 
H. B.SANTEE 
J. L. SPENCE 
H. M.STOLLER 
R. C. WILLMAN 
A. WISE 



E. I. SPONABLE 
R. O. STROCK 
S. K. WOLF 



C. STRUSS 
F. WALLER 



H. G. TASKER, Past-Chairman 
D. E. HYNDMAN, Sec.-Treas. 



H. GRIFFIN, Manager 
M. C. BATSEL, Manager 

PACIFIC COAST SECTION 



G. F. RACKETT, Chairman 



E. HUSE, Past- Chair man 
H. W. MOYSE, Sec.-Treas. 



W. C. HARCUS, Manager 
K. F. MORGAN, Manager 



SOCIETY ANNOUNCEMENTS 

CORRECTION TO PUBLISHED STANDARDS 

At the bottom of Chart 31, in Standards Adopted by the Society of Motion 
Picture Engineers, published in the November, 1934, issue of the JOURNAL, 
p. 283, an error appeared which should be corrected immediately. The sentence 
reading "Viewed from the light source, the sound-track is to the left" should be 
corrected to read "Viewed from the objective, the sound-track is to the left." 

This correction should be made also in all reprints of the standards that have 
been distributed. Copies of these reprints are available at the General Office of 
the Society at a cost of twenty-five cents each. 

ATLANTIC COAST SECTION 

At the regular monthly meeting of the Section, held on March 20th in the new 
re-recording building at the Vitaphone Studios of Warner Bros. Pictures, Inc., 
New York, papers were presented by Messrs. P. H. Evans and C. K. Wilson 
entitled, respectively, "Present Day Re-Recording Practices" and "Facilities for 
Present Day Re-Recording." 

The meeting was attended by approximately 250 members and guests, and 
opportunity was provided for thoroughly inspecting the new building and in- 
stallations. Preceding the meeting an informal dinner was held which was at- 
tended by approximately seventy members. 

MID-WEST SECTION 

The regular monthly meeting of the Section was held on April 4th at the 
Electrical Association, Chicago, 111. Mr. O. J. Holmes, of the Holmes Projector 
Corp., presented a paper on "New Developments in Portable Sound-Film Pro- 
jectors," illustrating his paper with demonstrations of the equipment. The 
meeting was well attended, and plans were laid for the next meeting of the Section, 
to be held on May 2. 

STANDARDS COMMITTEE 

Among the projects considered by the Standards Committee at its last meeting, 
held on March 28th, were those of possible standardization of sound, printer, 
and camera sprockets. Steps were taken to gather data on the subjects and to 
present these data at the next meeting. The question as to the approved method 
of guiding film in cameras (i. e., edge-guiding, sprocket-guiding, or a combi- 
nation of both) was answered by the fact that the Society had already approved 
edge-guiding by establishing all the dimensions of the film lay-outs in the vari- 
ous charts of the S. M. P. E. Standards from a designated guided edge. The 
adoption or recommendation of methods of determining screen brightness was 

463 



464 SOCIBTY ANNOUNCEMENTS 

tabled pending anticipated reports of the Projection Screens Committee and the 
Projection Practice Committee. 

The problem facing the Society, and, in fact, the entire American 16-mm. 
sound-film industry, was thoroughly discussed by the Committee, and it was felt 
that steps should be taken to establish the S. M. P. E. 16-mm. sound-film standards 
as American national standards by having the newly formed Sectional Committee 
on Motion Pictures (A. S. A.) consider them for submission to the American 
Standards Association. 

Accordingly, the following resolution was adopted: 

"Resolved that the standards of the Society of Motion Picture Engineers 
published in the November, 1934, issue of the S. M. P. E. JOURNAL, approved by 
the Committee and subsequently by the Board of Governors of the Society, be 
submitted to the Sectional Committee on Motion Pictures, A. S. A., for subse- 
quent action by the American Standards Association, with the following correc- 
tion: 

"That the wording at the bottom of Chart 31 be changed from 'Viewed from 
the light source, the sound-track is to the left' to 'Viewed from the objective, 
the sound-track is to the left.' " 

Other subjects considered by the Committee, but held in abeyance for further 
action, were as follows: 8-mm. sound-film; British Standard Specification for 
photoelectric cells used in sound-film apparatus; a possible 16-mm. standard 
test-film, similar to the present 35-mm. test-film; and the possible standardiza- 
tion of recorder slit width and reel dimensions. 

SECTIONAL COMMITTEE ON MOTION PICTURES UNDER THE A. S. A. 

At the first meeting of the Sectional Committee, held on March 29, the 
resolution given above, under the heading "Standards Committee," was received 
and considered. In view of the fact that several additions are expected to be 
made to the membership of the Sectional Committee, from organizations that are 
particularly interested in 35-mm. standardization, it was decided to take action 
only upon that part of the project relating to 16-mm. standardization especially 
in view of the fact that the present discussion between American and European 
interests hinged only upon the 16-mm. standards. The Committee accordingly 
voted unanimously to approve the 16-mm. standards, and letter-ballots were 
subsequently mailed to all members of the Sectional Committee whether they 
were present at the meeting or not. Passage of the standards will be determined 
by the letter-ballots. Action upon the 35-mm. standards was withheld until the 
next meeting of the Sectional Committee. 

Plans were laid for designating delegates to discuss the 16-mm. project at the 
International Film Congress at Berlin, April 25th, and at the International 
Congress for Photography at Paris, in July. 



SPRING, 1935, CONVENTION 

SOCIETY OF MOTION PICTURE ENGINEERS 

HOTEL ROOSEVELT, HOLLYWOOD, CALIF. 

MAY 20-24, INCL. 
Officers and Committees in Charge 

PROGRAM AND FACILITIES 

W. C. KUNZMANN, Convention Vice-President 
J. I. CRABTREE, Editorial Vice-President 
]. O. BAKER, Chairman, Papers Committee 

LOCAL ARRANGEMENTS AND RECEPTION COMMITTEE 

P. MOLE, Chairman 

G.S.MITCHELL E. HUSE G. F. RACKETT 

W. J. QUINLAN K. F. MORGAN H. W. MOYSE 

J. A. BALL W. G. HARCUS J. A. DUBRAY 

C. W. HANDLEY E. C. RICHARDSON F. E. JAMES 

R. H. McCuLLOUGH R. G. LINDERMAN C. DREHER 

N. LEVINSON 

PROJECTION COMMITTEE 

H. GRIFFIN, Chairman 

J. O. AALBERG R. H. MCCULLOUGH 

L. E. CLARK K. F. MORGAN 

Officers and Members of Los Angeles Local 150, I. A. T. S. E. 

STUDIO AND NEW EQUIPMENT EXHIBIT 

O. F. NEU, Chairman 

H. GRIFFIN J. FRANK, JR. 

P. MOLE S. HARRIS 

BANQUET 

W. C. KUNZMANN, Chairman 

P. MOLE W. J. QUINLAN E. HUSB 

G. S. MITCHELL G. F. RACKETT O. F. NEU 

PUBLICITY COMMITTEE 

W. WHITMORE, Chairman 

J. J. FINN J. R. CAMERON A. JONES 

F. H. RICHARDSON G. E. MATTHEWS P. A. McGuiRE 

465 



466 SPRING, 1935, CONVENTION [J. s. M. P. E. 

MEMBERSHIP 

O. M. GLUNT, Financial Vice-P resident 

E. R. GEIB, Chairman, Membership Committee 

LADIES' RECEPTION COMMITTEE 

MRS. E. HUSE, Hostess 

assisted by 

MRS. G. F. RACKETT MRS. F. E. JAMES MRS. E. C. RICHARDSON 

MRS. W. J. QUINLAN MRS. P. MOLE MRS. F. C. COATES 

MRS. C. W. HANDLEY 

Headquarters 

The headquarters of the Convention will be the Hotel Roosevelt, where excellent 
accommodations and Convention facilities are assured. Registration will begin 
at 9 a.m. Monday, May 20th. A special suite will be provided for the ladies at- 
tending the convention. Rates for S. M. P. E. delegates, European plan, will be as 
follows : 

Single: $2.50 per day; one person, single bed. 

Double: $3. 50 per day; two persons, double bed. 

Double: $4.50 per day; two persons, twin beds. 

Suites: $6.00 and $8.00 per day. 

Technical Sessions 

An attractive program of technical papers and presentations follows. Several 
sessions will be held in the evening, to permit those to attend who would be 
otherwise engaged in the daytime. All sessions will be held at the Hotel. 

Studio and Equipment Exhibit 

The exhibit at this Convention will feature apparatus and equipment developed 
in the studios, in addition to the usual commercial equipment. All studios are 
urged to participate by exhibiting any particular equipment or devices they may 
have constructed or devised to suit their individual problems, conform to their 
particular operating conditions, or to achieve economies in production, facilitate 
their work, or improve their products. 

Those desiring to participate should communicate with the General Office of 
the Society, Hotel Pennsylvania, New York, N. Y. No charge will be made for 
space. Each exhibitor should display a card carrying the name of the particular 
studio or manufacturer, and each piece of equipment should be plainly labelled. 
In addition, an expert should be in attendance who is capable of explaining the 
technical features of the exhibit to the Convention delegates. Expenses inci- 
dental to installing and removing equipment, and the cost of any power consumed, 
will be borne by the exhibitors. 

Semi-Annual Banquet 

The semi-annual banquet of the Society will be held at the Hotel on Wednesday, 
May 22nd. Addresses will be delivered by prominent members of the industry, 
followed by dancing and entertainment. Tables reserved for 8, 10, or 12 persons; 
tickets obtainable at the registration desk. 



May, 1935 J SPRING, 1935, CONVENTION 467 

Studio Visits 

S. M. P. E. delegates to the Convention have been courteously granted the 
privilege of visiting and inspecting the Warner Bros. First National Studio 
(courtesy of the Electrical Dept.), the Fox Hill Studio of Fox Film Corp., and the 
Walt Disney Studio: admission by registration card only. A visit has also been 
arranged to the California Institute of Technology. All bus charges on studio and 
other trips will be assumed by the individual delegates. 

Motion Pictures 

Passes will be available during the Convention to those registering, to Grau- 
man's Chinese and Egyptian Theaters, Pantages' Hollywood Theater, Warner 
Bros.' Hollywood Theater, and Gore Bros.' Iris Theater. 

Ladies' Program 

An especially attractive program for the ladies attending the Convention is 
being arranged by Mrs. E. Huse, hostess, and her Ladies' Committee. A suite 
will be provided in the Hotel where the ladies will register and meet for the various 
events upon their program. 

Further details of the Convention will be published in the next issue of the 
JOURNAL. 

Points of Interest 

En route: the gigantic Boulder Dam project, Las Vegas, Nevada; Union 
Pacific Railroad. 

Hollywood and vicinity: Beautiful Catalina Island; Zeiss Planetarium (Open 
May 1st); Mt. Wilson Observatory; Lookout Point, on Lookout Mountain; 
Huntington Library and Art Gallery (by appointment only) ; Palm Springs, Calif. ; 
beaches at Ocean Park and Venice, Calif.; famous old Spanish missions; Los 
Angeles Museum (housing the S. M. P. E. motion picture exhibit) ; Mexican village 
and street, Los Angeles; the California Pacific International Exposition at San 
Diego, Calif, (open May 29th) ; Agua Calienta, Mexico; Tia Juana, Mexico. 

Golf 

Members attending the Convention have been extended the privileges, at the 
usual course rates, of the following courses: 
Hollywood Country Club, North Hollywood 
Oakmont Country Club, Glendale 
Westwood Country Club, Westwood 
Rancho Golf Club, Westwood 



468 SPRING, 1935, CONVENTION [J. S. M. P. E. 

TENTATIVE PROGRAM 

OF THE 
SPRING CONVENTION 

HOTEL ROOSEVELT, HOLLYWOOD, CALIF. 
MAY 20-24, 1935 

MONDAY, MAY 20TH 

9:00 a.m. Registration. 
10:00 a.m. General Session. 
Address of Welcome. 
Presidential Response; H. G. Tasker. 
Society Business. 

Report of Membership Committee; E. R. Geib, Chairman. 
Report of Progress Committee; J. G. Frayne, Chairman. 

Report of Non-Theatrical Equipment Committee; R. F. Mitchell, Chairman. 
"Non-Theatrical Projection"; R. F. Mitchell, Bell & Howell Co., Chicago, 111. 
"Televison and Motion Pictures"; A. N. Goldsmith, New York, N. Y. 
"The Talking Book"; J. O. Kleber, American Foundation for the Blind, New 

York, N. Y. 

"Use of Films and Motion Picture Equipment in Schools"; Miss M. Evans, 
San Diego City Schools, San Diego, Calif. 

12:30 p.m. Informal Get-Together Luncheon. 

For members and guests of the Society; speakers to be announced later. 
2:00 p.m. General Session. 

Report of the Historical Committee; W. E. Theisen, Chairman. 

"A Description of the Historical Motion Picture Exhibit in the Los Angeles 
Museum"; W. E. Theisen, Honorary Curator, Motion Picture and Theatrical 
Arts Section, Los Angeles Museum, Los Angeles, Calif. 

"The Kodachrome Process of Amateur Cinematography in Natural Color"; 
L. Mannes and L. Godowsky, Eastman Kodak Company, Rochester, N. Y. 

"Introduction to the Photographic Possibilities of Polarized Light"; F. W. 
Tuttle and J. W. McFarlane, Eastman Kodak Company, Rochester, N. Y. 

"Production Problems of the Writer Related to the Technician"; C. Wilson, 
Metro-Goldwyn-Mayer Studios, Culver City, Calif. 

"Production Problems of the Actor Related to the Technician"; D, C. Jen- 
nings, Hollywood, Calif. 

"The Inter-Relation of the Dramatic and Technical Aspects of Motion Pic- 
tures"; Prof. B. V. Morkovin, University of Southern California, Los 
Angeles, Calif. 

"The Problems of a Motion Picture Research Library"; Miss H. G. Percey, 
Paramount Productions, Inc., Hollywood, Calif. 

8:00 p.m. Studio Visit. 

Visit to Walt Disney Studio, under the direction of Mr. W. Garity, Studio 
Manager; admission by registration card only; buses leave the Hotel 
promptly at 7:30 p.m. 



May, 1935] SPRING, 1935, CONVENTION 469 

TUESDAY, MAY 2 1ST 

9:30 a.m. Studio Session. 

Report of the Committee on Standards and Nomenclature; E. K. Carver, 
Chairman. 

"Process Cinematography" ; J. A. Norling, Loucks & Norling, New York, N. Y. 

"Calibrated Multi-Frequency Test Film"; F. C. Gilbert, Electrical Research 
Products, Inc., New York, N. Y. 

"Some Background Considerations of Sound System Service"; J. S. Ward, 
Electrical Research Products, Inc., New York, N. Y. 

"Modern Methods of Servicing Sound Motion Picture Equipment"; C. C. 
Aiken, RCA Manufacturing Company, Camden, N. J. 

"Technic of Present-Day Motion Picture Photography"; V. E. Miller, Para- 
mount Studios, Hollywood, Calif. 

"Engineering Technic in Pre-Editing Motion Pictures"; M. J. Abbott, RKO 
Studios, Hollywood, Calif. 

"The Analysis of Harmonic Distortion in a Photographic Sound by Means of an 
Electrical Frequency Analyzer"; O. Sandvik, V. C. Hall, and W. K. Grim- 
wood, Eastman Kodak Company, Rochester, N. Y. 

"Make-Up for Motion Pictures"; M. Firestine, Max Factor, Inc., Hollywood, 
Calif. 

1 :30 p.m. Luncheon and Studio Visit. 

Luncheon on the lot, and inspection of Warner Bros. First National Studio; 
courtesy of the Electrical Department, under the direction of Mr. F. Murphy, 
Chief Studio Engineer. Admission by registration card only; buses leave the 
Hotel promptly at 1:00 p.m. 

8:00 p.m. Meeting of the Technicians Branch of the Academy of Motion 

Picture Arts and Sciences; Mr. K. MacGowan, presiding. Members and 
guests of the S. M. P. E. are cordially invited. 

"The Technicolor Process"; J. A. Ball, Technicolor Motion Picture Corpora- 
tion, Hollywood, Calif. 

"Psychology of Color"; Natalie Kalmus, Technicolor Motion Picture Cor- 
poration, Hollywood, Calif. 

"Some Problems in Directing Color Motion Pictures"; R. Mamoulian, Holly- 
wood, Calif. 

Feature Motion Picture in Color: Becky Sharp. 

WEDNESDAY, MAY 22ND 

9:30 a.m. Laboratory Session. 

"The Argentometer an Apparatus for Testing for Silver in a Fixing Bath"; 

W. Weyerts and K. C. D. Hickman, Eastman Kodak Company, Rochester, 

N.Y. 
"Motion Picture Film Processing Laboratories in Great Britain"; I. D. Wrat- 

ten, Kodak Limited, London, England. 
"A Continuous Printer for Optically Reducing a Sound Record from 35-Mm. 

to 16-Mm. Film"; O. Sandvik, Eastman Kodak Company, Rochester, 

N.Y. 



470 SPRING, 1935, CONVENTION [J. S. M. P. E. 

"Optical Printing"; L. Dunn, RKO Studios, Hollywood, Calif. 
"Non-Uniformity in Photographic Development"; J. Crabtree, Bell Telephone 

Laboratories, Inc., New York, N. Y. 
"A Dynamic Check on the Processing of Film for Sound Records"; F. G. 

Albin, United Artists Studios, Hollywood, Calif. 
"New Agfa Motion Picture Film Types"; W. Leahy, Agfa Ansco Corporation, 

Hollywood, Calif. 
"Some Sensitometric Studies of Hollywood Laboratory Conditions" ; H. Meyer, 

Agfa Ansco Corporation, Hollywood, Calif. 

2:30 p.m. Studio Visit. 

A Visit to the Fox Hill Studio, under the direction of Mr. W. J. Quinlan, Chief 
Studio Engineer. Admission by registration card only; buses leave the 
Hotel promptly at 2:00 p.m. 

7:30 p.m. Semi- Annual S. M. P. E. Banquet. 

The semi-annual banquet and dance of the Society will be held in the New 
Supper Room of the Hotel. Addresses by eminent members of the motion 
picture industry. Tables reserved at the registration desk, for 8, 10, and 12 
persons. 

THURSDAY, MAY 23RD 

9:30 a.m. Projection and Studio Lighting Session. 
Report of the Projection Practice Committee; J. O. Baker, Chairman. 
Report of the Projection Screen Brightness Committee; C. Tuttle, Chairman. 
"The Relation between Projector Illumination and Screen Size"; D. Lyman, 

Eastman Kodak Company, Rochester, N. Y. 
"The Optical Efficiency of Mirror Guards"; W. B. Rayton, Bausch & Lomb 

Optical Company, Rochester, N. Y. 

"The Photoelectric Cell and Its Use in Sound Motion Pictures"; M. F. Jame- 
son, Bell Telephone Laboratories, Inc., New York, N. Y. 
Report of the Studio Lighting Committee; R. E. Farnham, Chairman. 
"The Radiant Energy Delivered on Motion Picture Sets from Carbon Arc 

Studio Light Sources"; F. T. Bowditch and A. C. Downes, National Carbon 

Company, Cleveland, Ohio. 
"The Photographic Effectiveness of Carbon Arc Studio Light Sources"; F. T. 

Bowditch and A. C. Downes, National Carbon Company, Cleveland, Ohio. 
"Lighting for Technicolor Motion Pictures" ; C. W. Handley, National Carbon 

Company, Los Angles, Calif. 
"A New Wide-Range Spot Lamp"; E. C. Richardson, Mole-Richardson, Inc., 

Hollywood, Calif. 
"Sources of Direct Current for Non-Rotating High-Intensity Reflect-Arc 

Lamps"; C. C. Dash, Hertner Electric Company, Cleveland, Ohio. 

2:00 p.m. Sound and Standardization Session. 

Interim Reports of Academy Committees on the Release Print and Screen 
Brightness; G. S. Mitchell, Manager, Research Council, Academy of Mo- 
tion Picture Arts and Sciences. 



May, 1935] SPRING, 1935, CONVENTION 471 

"The Technical Aspects of Recording Music for Motion Pictures"; R. H. 

Townsend, Fox Film Corporation, Hollywood, Calif. 
"A Device for Automatically Controlling the Balance between Recorded 

Sounds"; W. A. Mueller, Warner Bros. First National, Burbank, Calif. 
"Improvements in Play-Back Disk Recording"; G. M. Best, Warner Bros. 

First National, Burbank, Calif. 
"The Projection Background Process"; F. Jackman, Warner Bros. First 

National, Burbank, Calif. 

2:30 p.m. California Institute of Technology. 

A visit to the Institute, under the direction of Dean F. W. Hinrichs, Jr.; 
inspection of the astronomical, aeronautic, and high- voltage laboratories. 
Admission by registration card only; buses leave the Hotel for Pasadena 
promptly at 1 : 30 p.m. a beautiful scenic trip. 

8:00 p.m. Studio Session. 

Report of the Sound Committee; P. H. Evans, Chairman. 

"Newsreel Standardization"; J. A. Battle, Electrical Research Products, Inc. 
New York, N. Y. 

"Non-Directional Moving-Coil Microphone"; F. F,. Romanow and R. N. 
Marshall, Bell Telephone Laboratories, Inc., New York, N. Y. 

"Wide-Range Reproduction in Theaters"; J. P. Maxfield and C. Flannagan, 
Electrical Research Products, Inc., New York, N. Y. 

"Optical Printing of 35-Mm. Sound Records"; G. L. Dimmick, RCA Manu- 
facturing Company, Camden, N. J. 

FRIDAY, MAY 24TH 

9:30 a.m. Sound and Acoustics Session. 
"Sixteen-Mm. Negative- Positive and Grain"; D. Norwood, Lt., U. S. Army 

Air Corps, Chanute Field, Rantoul, 111. 
"Modern Instruments for Acoustical Studies"; E. C. Wente, Bell Telephone 

Laboratories, Inc., New York, N. Y. 
"Principles of Measurement of Room Acoustics" ; E. C. Wente, Bell Telephone 

Laboratories, Inc., New York, N. Y. 

"Recent Developments in Architectural Acoustics"; V. O. Knudsen, Univer- 
sity of California, Los Angeles, Calif. 
"Studio Acoustics"; M. Rettinger, Pacific Insulation Company, Los Angeles, 

Calif. 
"Technical Considerations of the High-Fidelity Reproducer"; E. D. Cook, 

RCA Manufacturing Company, Camden, N. J. 
"Development and Design of the High-Fidelity Reproducer"; F. J. Loomis 

and E. W. Reynolds, RCA Manufacturing Company, Camden, N. J. 

2:00 p.m. General Session. 

"Technical Aspects of the Motion Picture" ; A. N. Goldsmith, New York, N. Y. 
"The Contribution of Dr. Lee deForest to the Electronic and Motion Picture 

Arts"; G. A. Chambers, Eastman Kodak Company, Hollywood, Calif. 
"The History of the Talking Picture"; W. E. Theisen, Hollywood, Calif. 



472 SPRING, 1935, CONVENTION 

Apparatus Symposium 

"Three New Kodascopes" ; N. Green, Eastman Kodak Company, Rochester, 

N. Y. 
"A Continuous Film Camera for High-Speed Photography"; C. T. Burke, 

General Radio Company, Cambridge, Mass. 
"A Professional 16-Mm. Projector with Intermittent Sprocket"; H. A. 

DeVry, Herman A. DeVry, Inc., Chicago, 111. 
"Arc Supply Generator for Use with Suprex Carbons"; O. S. Imes, Century 

Electric Company, St. Louis, Mo. 
"The Akers 35-Mm. Hand Camera"; W. Blunel, Akers Camera Company, 

Hollywood, Calif. 

"A Sound Reduction Printer"; O. B. Depue, Chicago, 111. 
"A 35-Mm. Automatic Daylight Sound Motion Picture Projector"; A. B. 

Scott, SCK Corporation, Hollywood, Calif. 
"Vitachrome Diffusionlite System and Lamps, Their Uses and Applications"; 

A. C. Jenkins, Vitachrome, Inc., Los Angeles, Calif. 
"The Use of Cinematography in Aircraft Flight Testing"; F. H. Colbohm, 

Douglas Aircraft Company, Inc., Santa Monica, Calif. 
"The Use of Motion Pictures for Human Power Measurements"; J. M. 

Albert, Chas. E. Bedaux Company, San Francisco, Calif. 
"The Motion Picture in Japan"; Y. Osawa, J. Osawa and Company, Ltd., 

Kyoto, Japan. 
"The Motion Picture Industry in India"; G. D. Lai, Delhi, India. 

8:00 p.m. Sound Session. 
"Recording Music for Motion Pictures"; M. C. Batsel, RCA Manufacturing 

Company, Camden, N. J. 
"Analysis of the Distortion Resulting from Sprocket-Hole Modulation"; E. W. 

Kellogg, RCA Manufacturing Company, Camden, N. J. 
"A Comparison of Variable- Density and Variable-Width Sound Records"; 

E. W. Kellogg, RCA Manufacturing Company, Camden, N. J. 
"A Consideration of Some Special Methods of Re-Recording"; E. D. Cook, 

RCA Manufacturing Company, Camden, N. J. 
"Characteristics of the Photophone Light-Modulating System"; L. T. Sacht- 

leben, RCA Manufacturing Company, Camden, N. J. 
"Mechanographic Recording of Motion Picture Sound-Track"; J. A. Miller, 

Miller Film, Inc., New York, N. Y. 
"Application of Vertical-Cut Recording to Sound Pictures"; K. F. Morgan, 

Electrical Research Products, Inc., Hollywood, Calif. 

This is a tentative program and, as such, is subject to change. The Society is 
not responsible for statements made by the authors. 

The expenses incident to trips to studios and other points of interest in connec- 
tion with this program are to be borne by the individual members. 



I JOURNAL 

OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 

Volume XXIV JUNE, 1935 Number 6 



CONTENTS 

Page 

New Developments in Micro Motion Picture Technic 

H. ROGER 475 

Recent Developments in the Use of Mazda Lamps for Color 
Motion Picture Photography R. E. FARNHAM 487 

A New Method of Improving the Frequency Characteristic of 
a Single-Ribbon Light Modulator A. F. CHORINE 493 

Some Factors in Photographic Sensitivity S. E. SHEPPARD 500 

The Use of the Motion Picture for Visual Instruction in the 
Public Schools of New York R. HOCHHEIMER 519 

Need for Uniform Density in Variable-Density Sound-Tracks . . 

F. H. RICHARDSON 524 

Rulings of the U. S. Supreme Court in Recent Patent Cases of 
the American Tri-Ergon Corporation 529 

Book Reviews 551 

Society Announcements 553 

Author Index, Vol. XXIV, January to June, 1935 555 

Classified Index, Vol. XXIV, January to June, 1935 557 



JOURNAL 

OF THE SOCIETY OF 

MOTION PICTURE ENGINEERS 



SYLVAN HARRIS, EDITOR 

Board of Editors 
J. I. CRABTREB, Chairman 

O. M. GLUNT A. C. HARDY L. A. JONES 

J. O. BAKER 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1935, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 



Officers of the Society 

President: HOMER G. TASKER, 4139 38th St., Long Island City, N. Y. 
Past-President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
Executive Vice-President: EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, 

Calif. 

Engineering Vice-President: LOYD A. JONES, Kodak Park, Rochester, N. Y. 
Editorial Vice-President: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Financial Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Convention Vice-President: WILLIAM C. KUNZMANN, Box 6087, Cleveland, Ohio. 
Secretary: JOHN H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York, N. Y. 

Governors 

MAX C. BATSEL, Front & Market Sts., Camden, N. J. 
LAWRENCE W. DAVEB, 250 W. 57th St., New York, N. Y. 
ARTHUR S. DICKINSON, 28 W. 44th St., New York, N. Y. 
HERBERT GRIFFIN, 90 Gold St., New York, N. Y. 
GERALD F. RACKETT, 823 N. Seward St., Hollywood, Calif. 
WILBUR B. RAYTON, 635 St. Paul St., Rochester, N. Y. 
SIDNEY K. WOLF, 250 W. 57th St., New York, N. Y. 



NEW DEVELOPMENTS IN MICRO MOTION PICTURE 

TECHNIC* 



H. ROGER** 

Summary. After a brief description of the various factors involved in designing 
and operating equipment for making micro motion pictures, apparatus developed by 
the author is described. This apparatus is the result of years of experience in such 
work in the biological laboratory, particularly in collaboration with Dr. Alexis Carrel 
during the past eleven years. The details of the apparatus are discussed under the 
headings: Optical Bench, Camera and Driving Mechanism, Motors, and Control 
Panel. A table is given showing the manner of presetting the equipment for any 
purpose of research. The paper concludes with a brief reference to the use of 16 -mm. 
equipment in micro- and macrocinematography. 

Those who have studied the history of the motion picture know 
that it originated in the physiological laboratory. The study of the 
movements of men and animals by such men as Muybridge in Amer- 
ica and Marey in France gave rise to the now existing enormous mo- 
tion picture industry. The fact that through its invention we have 
become masters of the elements of time was realized at an early stage 
of development. Unfortunately, for a long time scientists turned 
away from the motion picture, leaving it entirely to entertainment, 
not realizing its possibilities as an aid to research. Comparatively 
recently motion pictures have found their way back again into the 
research laboratory. 

Microscopic research, especially, has profited to a great extent, 
as many new facts about the structure and physical behavior of 
microorganisms (for example, blood and tissue cells of men and 
animals) have been revealed with the aid of the motion picture. 
Many outstanding research laboratories are today using the motion 
picture camera in one way or another for automatically recording 
microscopic or macroscopic phenomena. It should be mentioned 
here that a definite boundary line exists between research on the one 
side and education or teaching on the other in the use of the motion 

* Presented at the Fall, 1934, Meeting at New York, N. Y. 
** Rolab Photo-Science Laboratories, Sandy Hook, Conn. 

475 



476 H. ROGER [J. S. M. P. E. 

picture. Many film records obtained for research have never passed 
through a projector, their only use being the study, frame by frame, of 
minute changes in the specimen under observation. 

To the uninitiated an equipment for taking micro motion pictures 
appears to be practically identical to a photomicrographic outfit, 
the difference lying only in the camera. However, one appreciates 
the difference at once when the specimen to be photographed is taken 
into consideration. Generally speaking, the objects for photomicrog- 
raphy are either dead material, specially prepared for the purpose, 
stained to bring out details; or, if living, are used only for a brief 
instant, so as not to be affected by the light or heat. Mounting the 
specimen upon the ordinary glass slide, protected by a cover glass, is 
almost ideal from the optical view-point, because all lenses, eyepieces, 
and condensers are corrected for a certain standard of slide mounting. 

Two opposing factors must be taken into consideration when mo- 
tion pictures of living specimens are to be taken over a period of time. 
Biological conditions must be fulfilled in the first place; that is, the 
specimen must be prepared and maintained in such a way as to render 
living conditions as natural as practically possible. The specimen 
is very often extremely delicate and sensitive to light, heat, atmos- 
pheric, chemical, bacteriological, and other influences, and has to 
be protected. These facts make it difficult to fulfill the optical and 
photographic requirements, and the outcome is in the best case a 
compromise under which the biological factor requires first considera- 
tion. 

To illustrate some of the problems just mentioned the author's 
experiences with living cells of tissue and blood may be reported. 
The study of tissue cultures, as now conducted in the laboratories of 
well known institutions, here and abroad, is of the greatest impor- 
tance to the future welfare of humanity. It offers, in fact, a possi- 
bility of solving many medical problems, the most outstanding of 
which is that of cancer. Experiments with normal and malignant 
tissues, irradiated with x-rays, gamma rays, etc., have been and are 
being carried on and may lead to a final control of the disease. 

One of the outstanding pioneers of research who succeeded in cul- 
tivating tissue cells outside the body, and winner of the Nobel prize 
in medicine of 1912, is Dr. Alexis Carrel, with whom the author had 
the privilege of collaborating for the past eleven years. His now 
almost 23-year-old strain of living tissue, originating from a frag- 
ment of chick embryo heart which was extirpated on January 17, 



June, 1935] MlCRO MOTION PICTURE TECHNIC 



477 



1912, is still as active as at any time, indicating that tissue is immortal 
if kept in an environment normal to the cells. 

The author's problem was to study by means of motion pictures the 
structure and the physical behavior of cell groups and single cells, 
normal and malignant, their influence upon each other, and the 




FIG. 1. 



Apparatus for taking micro motion pictures, developed by the 
author. 



changes that take place under the influence of various substances or 
radiations, etc. 

In order better to understand the optical problems and the micro- 
scopic technic, and the arrangement of the entire experimental set-up, 
it is necessary to describe briefly the cultivation of living tissue 
and blood. The small fragments of tissue taken either from the 
animal or from another culture are placed into a culture flask about 
5 cm. in diameter and 8 mm. high having an oblique neck 8 mm. inside 



478 H. ROGER [j. s. M. p. E. 

diameter through which all manipulations are made with long in- 
struments, such as platinum rods and long pipettes. Needless to say, 
the strictest sterile methods are adhered to, as the slightest deviation 
from them will cause certain infection or poisoning of the specimen. 
Great numbers of these flasks are kept in incubators and opened at 
regular intervals in order to supply the cells with a nutrient medium 
and to wash off the waste produced by the cells. The flasks, which 
ordinarily are hermetically closed, can be placed under a microscope 
and examined at low power. For high-power microscopic observa- 
tion the ideal way would be to place the specimen upon an ordinary 
slide with cover glass. However, the manipulation as well as the 
lack of oxygen necessary for the maintenance of life would impair the 
conditions and cause deterioration. Special flasks are therefore now 
used which are blown with extremely thin walls so that high-power 
oil immersions can be employed. The image so obtained permits the 
study of minute details of the cell structure, although not all the opti- 
cal requirements are fulfilled. In high-power microscopy the so-called 
critical illumination of the object is just as important as the formation 
of the image through the system of magnifying lenses. With the 
use of these micro flasks, an air space about the culture is biologi- 
cally necessary, but is not desirable from the optical point of view. 
Although a special substage condenser with a long working distance 
brought about certain improvements, the ideal arrangement could 
not be employed, which would prevail if an ordinary slide were used 
with no air between the front lens of the objective and the upper 
surface of the condenser. Experiments with a special culture cham- 
ber, made of optical glass, were carried on some time ago, but the 
danger of infection was increased and its use for routine work, so im- 
portant with tissue research, was greatly limited. 

When viewed under the microscope the cultures seem to be mo- 
tionless, and it would be extremely tiresome, if not impossible, to 
detect any changes in position by observing the specimen for any 
length of time. Here the motion picture camera, in combination 
with the microscope and a timing instrument (or stop-motion de- 
vice), has found its place in the research laboratory. With it, it is 
possible to make automatic observations over any length of time, of 
the development and the growth of tissue cells as takes place during 
the healing of wounds, the division of cells, or in phagocytosis ; and 
at higher magnification, of structural details such as of the nucleus, 
nucleolus, granules, mitochondria, vesicles, etc. 



June, 1935] 



MICRO MOTION PICTURE TECHNIC 



479 



It might be mentioned here that with increasing magnification 
the angle of vision, and hence the apparent speed of the object, is 
magnified proportionately. This naturally calls for an increasing 
frequency of exposure and hence an increase of light intensity. 
Considering the damaging effect of strong light upon living cells the 
number of problems also increases with higher magnifications. 

Equipment for taking micro motion pictures, at low as well as at 
high magnifications, must be so arranged as to permit a great 




FIG. 2. Micro motion picture equipment for 16-mm. 
cameras; upper part turned 180 degrees; microscope on 
separate table. 

variation in taking speed. Table I indicates the most useful settings 
for any purpose of research. The table is self-explanatory. 

MICROCINEMATOGRAPHIC APPARATUS 

Great indeed is the number of authors in recent years who have 
described in scientific journals equipment for taking micro motion 
pictures. It can be said of most of them, however, that their set-ups 
are incomplete, serving only one particular purpose. On the other 
hand, apparatus developed and sold by a few manufacturers, although 



480 H. ROGER [j. s. M. p. E. 

the product of clever designing engineers, are practically useless for 
scientific research because their design is not the result of laboratory 
experience. All this seems to indicate that biologists are inexpert 
technicians and technicians are inept biologists. 

It is true that occasionally good work has been accomplished by 
skilled workers with equipment that is comparatively crude, but for 
continuous work and results consistently of a certain standard of 
quality, an apparatus must be available that can be operated by 
any one in the laboratory and the set-up of which can be altered to 
suit any purpose. The number of adjustments must be limited to 
a minimum and the operator's attention must mainly be focused upon 
the specimen under investigation. 

The Standard Microcinema Apparatus that will now be described 
is the result of a slow process of development, reaching back over 
some twenty years of laboratory experience. It was designed with 
the view in mind of making its operation as simple as possible without 
sacrificing quality of results, so important in research procedure. 

In its original form, the apparatus was designed for the purposes of 
the most exacting research, where magnification, resolution of detail, 
and exact timing are concerned, so important for drawing conclusions 
of value to the scientist. Every part of the instrument and its func- 
tion was carefully conceived and incorporated in its logical place. 
The number of manipulations is reduced to a minimum, making it 
possible for the experimenter to focus his attention upon the subject 
rather than upon its photography. 

In viewing the apparatus, a layman might perhaps wonder why is it 
necessary to employ such heavy equipment, weighing about 500 
pounds, to photograph images the size of which is only 18 by 24 mm. 

Considering, however, that, if an object is enlarged several hundred 
times its diameter, any motion of the object is also magnified at the 
same rate, it will then easily be understood that even the slightest 
vibration, not detectable by ordinary means, would exert a damaging 
effect upon the details and sharpness of the image. It is here a 
decided advantage to make the apparatus as heavy as possible. A 
saving in cast-iron, of which the apparatus is mainly constructed, 
would be negligible. The absorption of vibrations will be discussed 
later. 

The apparatus (Fig. 1) consists of four units, which are separate 
from each other and have no mechanical connections between them 
other than the floor upon which they stand, and some flexible material, 



June, 1935] MICRO MOTION PICTURE TECHNIC 481 

such as a leather bellows and an endless belt. The four units are 
as follows : 

(A) Optical Bench. This consists of a cast-iron stand which 
carries the microscope, the light source, condensers, cooling cells, 
color screens, and neutral screens, all mounted upon adjustable riders 
with the exception of the microscope, which stands upon an ad- 
justable table. The three legs are so arranged that their height can 
be altered according to the experimenter's convenience. Levelling 
screws under each leg rest upon vibration absorbers, which are 
intended to eliminate the transmission of all vibrations of the building 
to the microscope. The microscope table itself can be adjusted 
within certain limits in all directions in order to bring it into optical 
alignment with the camera and the central beam of light. 

(B) Camera and Driving Mechanism. A heavy cast-iron base 
and two vertical columns support a cross-member bearing an electric 
relay, a platform for the camera, focusing device, driving mechanism 
and clutch, and motor, equipped with limit switches for raising and 
lowering the platform. 

(C) Motors. The direct motor is equipped with a governor for 
controlling the speed; the indirect motor with reduction gears and an 
electromagnetic brake. 

(D) Control Panel. Receptacles are provided having capacities 
up to 30 amperes, voltmeters and ammeters, two dials for setting the 
time intervals, per hour or per minute (see Table I), and a time clock 
for presetting the start and stop up to 12 hours in advance. 

To increase the temperature, as is always necessary when working 
with tissue cultures and in bacteriological studies, the microscope is 
surrounded by an incubator which is so arranged that it can be re- 
moved and replaced in a few seconds without touching the micro- 
scope. The coarse and fine adjustment screws of the microscope 
have extensions to the outside of the incubator. Doors permit easy 
handling of the specimen and the stage of the microscope. All 
manipulations can be observed through a glass window in the upper 
part of the incubator. It is, in fact, as easy to use the microscope 
with the incubator in place as when used in the ordinary way upon a 
table. 

To exclude external light there must be a connection between the 
microscope and the camera, which, by the way, is used without a 
camera objective. On the Standard Microcinematographic Apparatus 
there are three attachments which can be used alternatively: namely, 



1 

6) 

I 



i 
1 









_ 




~S 








<N T)< TfN (N Tf< 




= |=0 | 




5 




5 




aijv 




O O OO 




SSoo^^S^^M^ 




O(N rH rH rH 












-H 




8OOOOCDOOOCO 
OOO OCO OOO CO 
-*< <N rH CD O 00 tO <N 




O lOOOIN i-H rH 
















03g 












IN <N CO 




o 




5 




!W 








O O OOOO 
**< (N T* (N ^ (N 




OOCOOOO lO CO (N 

TjH (N rH rH 




" 




* 




-H 




OOOOOOO 
^ CO <N 00 CD TfH C^ 




OOCOOCOOCOCOINrH 
00 * <N (N I-H rH 
















'O3S 








OOOO 




OOCOOO O 00 O 

<J<(NIO'*C^ IN rH T* 




5 




= 




uijM 




OO 
* (N rf*(N 




OOCDOCOOtOOCDOO 

(NTj(TjN>OTj<(NrHTf<lO 




COCOOOcOiO CO IN rH 












-H 




OOcOOOOOCOCO 




lO<NrHrH 
















-s 








OOOO 




\OT \OT 

OOCOOO O CO O 
<N * CN IN Tt< Tf iO 




8 




~ 




uipv 




O O OO 
(N -^ - (N rj< 




OOCOOCOOCOCOCOCO 
& (N >O * (N IN rH iO N 




CO CO^ 00 <N rH rH 






! 






-H 




OCO O CD OCO CD 
OCOOCDOCOCD 
00 lO Tj* (N (N rH 




CD CO 00 CO * CO CN rH 






^ 






d 

"o 




V 


o 3S 








O OOOO 
(N <* (N TjH (N T(* 




W\ H\ W\ iH\ 
OOCOlOO <N rH rH 
INrH^ ^ T* CO 


o 


-" 


b 


5 


;- 


mm 




5 5 




OOOOOCOiOOOr-icOCO 

-J1(NCOrHTj< (N-^COrH 




00^ (N (N rH rH 


s 

fc, 


CO 




3 






-H 




O OO O CO IO 00 rH 
O 00 *O CD CN 00 * 




CO 00 lO <* (N (N rH 
















o, S 








O O OiOOO 
T* (N 10<NTJ*<N 




OOCOOO O CO 10 

*CNrHT*(N 10 CO (N 









o 




mpn 




OO 
^ (N TfH (N 




OOCOOCOOCDCOCDCO 

(N TtN (N rH Tt< COCMrH 




CO 00 IN rH rH 












^H 




DOCOOOOOCOCO 




CO CO Tt< CO <N I-H rH 
















03S 








O O OOOO 
(N Tj< (NTjtiN'* 




%%%% s s g 











^ 




UIJV 




O O OO 
<M -<J< (N-* 




5co^SSS2 




IO(N rH rH rH 












JH 




OOOOOCOOCOCO 

TjN rH CO O 00 10 CN 

CO COIN rH rH 




O lOCO <N rH rH 
















33S 








OOOO 
T* (N TjNTj<<N 




OOOOOlN O OO O 
* <N 10 Tj< CO <N rH rH 




g 








UIK 




O 00 

T* (N "*<N 




OOCDOCOOCOCOO0 

(N -r)< TfH (N >0 * IN rH rH 




IN rH 












"TO 




CO rH rH 




iOCNrHrH 
















-o 3S 








O O OOOO 
(N CD (N ^(N -* 




oofoco o 5 o 









CO 




nijM 




O O OO 
C<l ^ CN ^ 




OOOOOCOOCOCDiOlN 

<*< <N >O ^ <N (N rH 




^ 












JH 




O COOCOOCO CO 
00 10 Tt< IN CN rH 




C<1 rH 










it Projection 







an 
ft 

IN 




CDOOOOrH^Ot^CO 




OOOOOOOOOO 1 * 
^(NXCD^OOCNCOTjfiN 

^ t> Tt< 00 (N rH rH 




g "O U fl 

alo 






Acceleration ! 


yT 




en 

d 

CO 




oooooooo 
oooooooo 




OOOOOOOOCNCO 
CO 00 (N * CO <N 00 T!< CO rn 

O5 ^* 00 (N rH rH 




o o' 
x Tt< IN <N -H r- d-^-a^ 
ati v v .* 




I 


M 


, 




m 








gg8SS ^- 




XWK5 * J * 






F^ 


1 




^a 




gg22-"- 
















aj 


IIS 
J" 


odxg 

3^B^ 


J6| 


rH IN 00 * CO 00 C< * 


ill 


"^"*"2?;g 


H 


MTt<CDOOO O <* (N 

rH rH CS CO 



MICRO MOTION PICTURE TECHNIC 



483 



a flexible bellows, a telescope sleeve, and an observation eyepiece 
(beam-splitter). For macroscopic work, a small camera attachment 
can be used instead of the microscope, permitting the photography 
of small objects such as insects and other gross specimens. These 
can be placed conveniently upon the microscope table. 

The focusing device previously mentioned is mounted, together 
with the camera, upon the camera support, in such a way as to permit 
either the camera or the focusing device to be brought into alignment 




FIG. 3. 



Micro motion picture equipment used for 
macrophotography. 



with the optical axis of the microscope. This arrangement has 
proved through many years of practice to be very valuable in ob- 
taining critical sharpness of the images in the plane of the photo- 
graphic emulsion at the highest magnifications. 

With regard to the observation eyepiece (beam-splitter) it should 
be said here that all instruments of this kind, placed upon the market 
by a number of manufacturers and described frequently in the scien- 
tific literature, have only a limited use. The intensity of the light- 
source must be varied to a great extent, depending upon the mag- 



484 



H. ROGER 



[J. S. M. P. E. 



nification and the rate of exposure. Consequently, it would also be 
necessary to have a beam-splitter with a variable rate of reflection 
in order to compensate for the great changes of intensity, and so make 
it possible for the observer to see an image of approximately the same 
brightness in all cases. On the other hand, with all work requiring 
time-lapse photography there is generally enough time between ex- 
posures for checking the movement and the focus with the focusing 
device mentioned above, so that a beam-splitting device becomes 




FIG. 4. Stand and 16-mm. camera in combination with portable timing 
device for time-lapse work. 

unnecessary. Most of the research upon living cells requires time- 
lapse photography. 

An outstanding feature of the Standard Microcinematographic Ap- 
paratus is the possibility of quickly changing the speed (or the 
frequency at which the pictures are taken). In microcinematog- 
raphy the variety of problems, and thec hanges from lower to higher 
magnification and vice versa, necessitate changes in the rate of ex- 
posure and hence also in the light intensity, which often requires 
replacing one kind of light source for another. In work with appa- 
ratus of the old type such changes always involved a great deal of 
experimentation consisting mainly of changing gears or pulleys in 
complicated devices, checking the rate of exposure with a stop-watch, 



June, 1935] MlCRO MOTION PICTURE TECHNIC 485 

readjusting the substage condenser, altering the light source, making 
a number of test exposures, developing and fixing, etc. The waste of 
time and often the deterioration of the specimen during the time 
adjustments were made, caused micro motion pictures to be con- 
sidered a difficult art calling for men of knowledge and experience. 
Although it is practically impossible to build a machine that can be 
as simply adjusted as a radio receiver, much has been accomplished 
in reducing the difficulties of manipulation. The new apparatus 
described here is constructed in such a way as to make it easy to 
switch from one speed to another. It can be run either continually 
at various speeds or intermittently, which in itself offers a number of 
advantages. 

The direct and the indirect motors (the one with speed reducer) are 
mounted at equal distances from the main driving pulley, making it 
easy to switch the driving belt from the one to the other. For time- 
lapse photography, the apparatus is set for intermittent operation; 
that is, the mechanism does not run during intervals between pic- 
tures. The number of pictures to be taken per minute or per hour is 
controlled by two dials upon the control panel, each of which is con- 
nected to a synchronous motor which closes and opens an electric 
circuit at the predetermined time interval. A relay included in the 
circuit starts and stops the motor that drives the mechanism and at 
the same time turns the light source on and off synchronously with the 
camera. Thus, only one revolution is made and one picture taken. 
The advantages of such an arrangement are obvious, considering that 
the change from one speed to another is accomplished merely by 
setting the dial on the panel. The correct exposure, once found, al- 
ways remains the same whether the dial is set for one, two, three, or 
eight exposures per minute or per hour. 

The procedure of making a film record of a specimen is as follows : 

The specimen is placed upon the stage of the microscope in the 
usual manner, after which it is focused through the eyepiece of the 
focusing device. The distribution of light can be checked by switch- 
ing over a small ground-glass disk and adjusting the mirror of the 
microscope, if necessary. By lifting a handle at the front of the ap- 
paratus and shifting it to the right, the camera is brought into the 
taking position. A clutch connecting the driving mechanism with 
the camera operates automatically at the same time. By turning 
the main switch the apparatus is set into operation. 

For making test exposures or for loading new film, the camera 



H. ROGER 

does not need to be removed from the stand but can be tipped over 
and opened in the usual manner. 

[As a demonstration of the work done with the apparatus here described, a 
film was projected showing some of the results of research carried on at the 
Rockefeller Institute, New York, N. Y., and at the Rolab Photo-Science Labora- 
tories, Sandy Hook, Conn. The subjects were: living cells of tissue and blood, 
cell division, phagocytosis, details of cell structure, undulating membrane of 
white blood corpuscles (discovered with the aid of motion pictures), blood cir- 
culation, the growth of a culture of bacteria, Brownian movement of ultra- 
microscopic particles, and other physico-chemical subjects. Ed. ] 

PORTABLE EQUIPMENT FOR 16-MM. CAMERAS 

A camera stand and focusing device for 16-mm. cameras, having 
some of the features of the standard apparatus, has been described 
previously. 1 This stand with the additional features to be mentioned 
below, is especially designed for scientists and laboratories having 
limited resources but who may possess 16-mm. motion picture 
cameras, of any make. It is possible to do a great variety of work 
with the stand alone, the attachments and additional equipment being 
purchased later when required. The following combinations 
have been used successfully for a number of years in various labora- 
tories where these 16-mm. outfits are in operation: 

(A) Stand alone, with microscope and light source for slow- 
motion, normal speed, and stop-motion photography at low and high 
magnification. 

(B) Stand with upper part turned 180 degrees permitting the 
microscope to be placed upon a separate table (Fig. 2). 

(C) Stand with bellows attachment and objective for macrophotog- 
raphy (no microscope necessary); position normal, or upper part 
turned 180 degrees, permitting a greater distance between object 
and lens (Fig. 3). 

(D) Stand and camera in combination with portable timing de- 
vice for time-lapse work (Fig. 4). 

REFERENCE 

1 ROSENBERGER, H. : "Progress in Micro Cinematography," /. Soc. Mot. Pict. 
Eng., XV (Oct., 1930), No. 4, p. 439. 



RECENT DEVELOPMENTS IN THE USE OF MAZDA LAMPS 
FOR COLOR MOTION PICTURE PHOTOGRAPHY* 

R. E. FARNHAM** 

Summary. This paper supplements a previous one in which the advantages of 
operating tungsten filament lamps at very high efficiencies were explained. The 
paper discusses the development of a glass filter, which when used with 33-lumen-per- 
watt lamps gives practically perfect white light, suitable for color photography. 

In a previous paper 1 attention was called to the need of substanti- 
ally the same quantity of blue radiation from light sources as of green 
and red, made necessary by the inclusion of the third primary color, 
blue, in the newly available three-color processes of motion picture 
photography. The paper then discussed a method of approximately 
fulfilling this requirement with incandescent lamps operating at ex- 
tremely high efficiencies. 

Fig. 1 shows graphically the effect of higher efficiency operation. 
The horizontal scale indicates the wavelength or color of the radia- 
tion and the vertical scale the relative energy. The lower curve shows 
the character of the radiation from a standard 1000- or 1500- watt 
PS 52 bulb lamp now quite generally used for black-and-white photog- 
raphy. It will be observed that the radiation in the red region is 
about 4 J /2 times that in the blue-violet part of the spectrum. 

By increasing the efficiency of the lamp to 33 lumens per watt the 
over-abundance of red to blue- violet has decreased only 2.3 times. 
Or, another way of expressing it, the blue- violet has increased 200 per 
cent, while the red has increased only 60 per cent. The curves are 
for lamps of equal wattage. 

The actual effect is that the light of the high-efficiency lamp is 
very definitely whiter, and there is the added advantage of more 
total light for the same wattage since the lumen-per-watt output is 
greater. 

For the general lighting equipment a lamp of 2000- watt and 67,000- 

* Presented at the Fall, 1934, Meeting at New York, N.Y. 
** General Electric Company, Cleveland, Ohio. 

487 



488 



R. E. FARNHAM 



[J. S. M. P. E. 




4000 



7000 



5000 000 

Angstrom Units. 

FIG. 1. Spectral energy distribution of Movieflood lamp in comparison with 
that of the more usual types. 



June, 1935] USE OF MAZDA LAMPS 489 

lumen rating in the PS 52 bulb was developed and has since been 
called the Movieflood. For the modelling lighting requirement it was 
found that by operating a 5000-watt lamp of 105-volt rating on the 
usual 118- to 120- volt circuits of the motion picture studios, its ef- 
ficiency rose to 33 lumens per watt, the same as the Movieflood lamp. 
This is possible because the efficiency of the 5-kw. lamp now used for 
studio work is 29.0 lumens per watt. 

The requirements of one of the color processes was satisfactorily 
met by lamps of this efficiency. This was not sufficient, however, 
for another color process which requires an illuminant possessing 
more nearly equal quantities of the three primaries. Since further 
correction of the light quality by increasing the efficiency was hardly 
practicable from the standpoint of shortened lamp-life, the use of 
some form of correcting filter was necessary. 

This filter can be either at the camera lens or incorporated in the 
lighting equipment. Further study of this problem during the past 
summer brought out the further requirement that it is frequently 
desirable to employ on a particular set more than one type of il- 
luminant, such as the white-flame arc and Mazda lamps, or incandes- 
cent lamps and daylight. The color of the light must therefore be 
corrected at the source. 

This requirement imposes a somewhat closer tolerance on the 
filter than if Mazda lamps were used alone, because some adjust- 
ment must be made for variations in the exposing light in printing 
the color positive film. No correction is possible, however, when the 
actors move from light of one color into that of another. An analysis 
of the filter characteristics showed that it must be one with a high 
transmission in the extreme blue- violet region and a decreasing trans- 
mission toward the red end of the spectrum. 

In the actual solution of the problem a number of shades of gelatins 
were chosen whose transmissions approximated that of the required 
filter. Then a group of persons in variously colored costumes and 
settings were photographed first by the light of the white-flame arc 
and then by the high efficiency Mazda lamp with each of the gelatin 
filters, changing back to the control source each time before a new 
filter was put in place. Examination of the finished pictures indi- 
cated very definitely the proper filter, which proved to be the Brigham 
shade No. 26. 

The transmission characteristics, as well as a sample of the gelatin, 
were turned over to the Corning Glass Works, who endeavored to 



R. E. FARNHAM 



|J. S. M. P. E. 




4000 



7000 



5000 COOO 

Angstrom Units. 

FIG. 2. Effect of Coming's Lunar White No. 570 filter upon the color quality 
of the light of the Movieflood lamp. 



June, 1935] USE OF MAZDA LAMPS 491 

duplicate this in glass as well as to introduce some heat-reducing 
characteristics as well. A sample of the resultant filter was then 
tried out in the same manner as previously described. 

Fortunately, this first attempt resulted in a filter that was slightly 
better than the gelatin, there being no observable difference between 
pictures photographed with the arc source, and the 33-lumen-per- 
watt lamps and the filter. This filter is known as Coming's Lunar 
White No. 570. Actual density measurements from sensitometric 
data taken at the time the color photographs were made show the 
light from this combination of lamp and filter to have 100 units of 
red, 102 units of green, and 99 units of blue, an almost perfect white. 

Fig. 2 shows again the spectral energy distribution of the high- 
efficiency lamp, the transmission characteristics of the filter, and 
finally, the two in combination. 

REFERENCE 

FARNHAM, R. E.: "The Use of Mazda Lamps for Color Photography," /. 
Soc. Mot. Pict. Eng., XXI (Aug., 1933), No. 2, p. 166. 

DISCUSSION 

PRESIDENT GOLDSMITH: Does the filter have any heat-absorbing characteris- 
tics? 

MR. FARNHAM: Some, but not as much as I should like it to have. 

MR. POPOVICI: What is the color temperature? 

MR. FARNHAM: About 3450 degrees. Tungsten melts at 3655, so we are not a 
great deal below the melting point of tungsten. 

MR. POPOVICI : If the voltage drops, the color temperature will also drop, to a 
certain extent. What is the relation of the color temperature to the filter? 

MR. FARNHAM: Apparently the blue end of the spectrum of an incandescent 
lamp drops off rather rapidly as we reduce the voltage, so it is essential that the 
lamp be operated up to true voltage. 

MR. POPOVICI: What is the limit of allowable voltage drop? 

MR. FARNHAM: I should say 5 per cent, plus or minus; 110 to 120 volts, aver- 
aging 115, is about a fair operating figure. A slightly closer voltage operation is 
called for than is usually the practice in black-and-white photography. 

MR. MILI: Tests recently made in our laboratory indicate that a 5 per cent 
drop in voltage will lower the color temperature of the illuminant by about 75. 
In other words, if the color temperature of a photoflood lamp at 115 volts is 
3500K., a drop to 110 volts will lower the temperature to about 3425K., and a 
rise to 120 volts will raise the temperature to 3575K. While such a change 
would not be very important in color photography, it would be better to have the 
voltage across the lamp terminals closely regulated. For this reason it would 
be worth while to use lamps in series with a very small rheostat and adjust for 



492 R. E. FARNHAM 

constant voltage, in order that the blue to red ratio in the luminous spectrum may 
not change during the course of a sitting or the shooting of a scene. 

MR. PALMER: It is evident that color photography must be done much more 
carefully than black-and-white. When photographing in color, the density of the 
negatives must be kept quite uniform throughout. Consequently, the lighting 
of a given person must be the same in different sets. Since we have to be so care- 
ful about measuring the light, there should be no particular difficulty about being 
a little more careful about the voltages. 

MR. POPOVICI : In measuring color, we have to correct the meter according to 
its sensitivity to the kind of light that is used, because the photometer reading 
may remain constant even though the voltage should drop a little without being 
noticed. 

PRESIDENT GOLDSMITH: Until recently the color motion picture companies 
fought shy of using or taking out-door scenes to any extent, for later theater 
presentation. They seemed to restrict themselves to studio material taken under 
very closely regulated conditions. Of course, that is a severe restriction, and 
sooner or later there must be plenty of out-door scenes upon the screen. If the 
light changes in an out-door scene as it does every time a cloud drifts across the 
sky, the natural effect will be to have it show upon the screen. I am not so sure 
that the artificially perfectly maintained color in a studio production of color 
motion pictures is either a desirable or a faithful presentation. If we walk 
around this lecture room, our apparent color changes as we get nearer the green- 
gray reflecting walls or the orange-yellow lamps. If a color photograph were to 
be made of us in this room, it should show such effects. 

MR. FARNHAM : When color first began to be used, the colors were usually quite 
brilliant, and you usually found yourself looking at a color picture, and the 
story was secondary. Now the color pictures seem to have the pastel shades and 
are very much lighter and actually more realistic, so you find yourself first looking 
at a story ; then you feel that there is something more natural about it. 

PRESIDENT GOLDSMITH: Quite right. The fact is that in real life bright colors, 
saturated colors, and primary colors are very much the exception. We deal 
entirely with degraded, secondary, and tertiary colors. I don't believe that a 
single pure color could be found in this room. 



A NEW METHOD OF IMPROVING THE FREQUENCY 

CHARACTERISTIC OF A SINGLE-RIBBON LIGHT 

MODULATOR* 

A. F. CHORINE** 



Summary. After discussing briefly some of the difficulties involved in equalizing 
in the conventional manner the frequency characteristic of the light modulator pre- 
viously described by the author, a process is described of equalizing by means of a 
small drop of oil placed at a chosen point upon the vibrating ribbon. By varying 
the size of the drop and its location upon the ribbon, a whole series of characteristic 
curves is afforded from which any desired one may be chosen. The advantages of 
such a system, as regards the frequency response, factor of safety for the ribbon, and 
amplification, are described. 

In sound recording it is very necessary that the amplitude be 
adequate and the frequency characteristic of the light modulator 
satisfactory. It is the purpose here to describe the development and 
improvements in the author's light modulator described in detail in 
previous issues of the Journal. 1 

With respect to the amplitude characteristic, it is difficult to pro- 
duce anything essentially new because the modulator represents the 
commonly known sensitive string type of galvanometer which 
satisfies the most particular requirements. Fig. 1 shows the ampli- 
tude curve obtained with this instrument under the most strenuous 
conditions. Fig. 2 shows the schematic arrangement of the modulator. 
The current carried by the ribbon was so great as to cause a pro- 
nounced lengthening of the ribbon and to introduce other distortional 
factors. The proportionality between the current and the amplitude 
of the ribbon changes slightly; so, for that reason, in the construc- 
tion of the instrument efforts were directed toward decreasing the 
loading of the ribbon by (a) working with great magnification; 
(6) applying doubling; (c) increasing the magnetic field density. 
In the latest construction these three modifications have made it 

* Received March 7, 1935. 

** Director, Central Laboratory, All-Union Electrical Trust, Leningrad, 
U. S. S. R. 

493 



494 



A. F. CHORINE 



[J. S. M. p. E. 



i 



0.75 

1 
i - 5 



I 



possible to use a very small deflection of the ribbon (maximum am- 
plitude not greater than 0.03 mm.), and the general amplitude char- 
acteristic was exceptionally good. 
The frequency characteristic in 
air is shown in Fig. 3. It is possible 
to work with such a characteristic 
only in combination with a filter 
designed to attenuate at the fre- 
quency of resonance. If the 
resonance frequency is within the 
recording band, or is fairly low, a 
filter that would attenuate the 
entire band, including the resonance 
peak, is used. Such systems are 
not convenient because they require 

extraordinary care in recording and 
FIG. 1. Amplitude characteristic of f ,< .,1 

., very accurate tuning of the ribbon 

the modulator. J 

to a given natural frequency. Be- 
sides, in the case of small optical magnification, great current density 
resulted which caused the ribbon to burn out when overmodulated. 



0.25 























/ 






















/ 




















/ 




















/ 




















/ 




















^ 






















/ 




















7 




















/ 




















/ 




















/ 






1 














/ 




















/ 






















/ 




















</ 






















2 






















90 170 230 33( 
Milliamperes 




JDtt 

FIG. 2. ( Upper) Schematic arrangement of first light modulator developed in the 
author's laboratory. (Lower) Details of modulator illustrated above. 



June, 1935] 



IMPROVING THE LIGHT MODULATOR 



495 



Such systems had to be discarded for practical reasons. It was 
found to be more practicable to use damping fluids as a means of 
improving the frequency characteristic of the modulator. Best 

10 



1 

0.8 

0.6 

0.4 

0.2 

0.1 

10 
FIG. 3. 



60 100 200 400 1000 4000 10000 

Frequency characteristic in air after taking steps to 
decrease the loading of the ribbon. 

results have been attained by using vaseline as the damping medium. 
For a duralumin ribbon 18 mm. long, having a cross-section 0.175 
by 0.008 mm., characteristics have been normally attained like the 

10 



1 

0.8 
0.6 

0.4 

0.2 
0.1 



10 60 100 400 1000 4000 10000 

FIG. 4. Normal frequency characteristic of a duralumin ribbon 
18 mm. long and 0.175 by 0.008 mm. in cross-section. 

one shown in Fig. 4. The heavy curve represents the frequency 
characteristic obtained in air, and the dotted curve with the ribbon 
completely immersed in oil. 



496 



A. F. CHORINE 



[J. S. M. P. E. 



Analyzing the curve, we find that to satisfy the present technical 
requirements, it is necessary to improve the curve by lengthening as 
much as possible its linear portion. Extensive experimental work 
showed that a very good characteristic could be achieved by the 
comparatively simple method of depositing small drops of vaseline 




saoo sou ?ooo *xx> s 



FIG. 5. Characteristic of the ribbon with a drop of oil placed one-third 
the distance from the end of the ribbon; natural frequency, 4000 cycles. 

oil upon the half, third, or quarter length of the ribbon. The curves 
thus obtained are shown in Figs. 5, 6, and 7. 

Fig. 5 is the curve of the same duralumin ribbon, 18 mm. long and 
0.175 by 0.008 mm. in cross-section, with a drop of vaseline oil (about 
2.5 mm. in diameter) placed one-third the distance from the end of 
the ribbon, the natural frequency of which was 4000 cycles. This 
curve shows a considerably wider frequency band as compared with 



FIG. 6. Characteristic o